人工智能赋能教学设计新范式

ABSTRACT AI-assisted learning is gaining recognition for its potential to enhance educational outcomes, yet its impact on student satisfaction, particularly in the context of programming courses, has not been thoroughly investigated. This study aimed to fill this gap by assessing satisfaction among 70 secondary students through a three-week AI-assisted project as part of their programming course. The results revealed a generally positive reception, with satisfaction levels consistent across genders, grades, and experiences with AI-assisted learning. Notably, a robust positive correlation was observed between satisfaction with human instructors and other components of the learning experience. In contrast, the learning sequence was found to have a counterintuitive negative correlation with overall satisfaction. Building on these insights, we developed a conceptual framework that delineates the factors determining student satisfaction in AI-assisted learning environments. Our findings underscore the need for a balanced approach that incorporates both AI technology and human interaction in educational settings.

Artificial Intelligence (AI) has become a powerful tool in English as a Foreign Language (EFL), offering significant prospects for improving language learning and teaching. Recently, the incorporation of chatbots, one of the advanced AI language models, in EFL writing has garnered interest. This study aims to investigate the use of AI chatbots in EFL writing instruction, driven by their potential to stimulate student engagement across affective, behavioral, and cognitive engagement. The main objective was to evaluate student engagement levels with AI chatbots and assess EFL teachers' strategies for stimulating this engagement. Utilizing a mixed-methods design, the research involved 40 students and two faculty members, employing questionnaires and semi-structured interviews for data collection. Quantitative data was analyzed using SPSS, and qualitative insights were obtained through thematic analysis of interview transcripts. Findings indicate that AI chatbots significantly improve student engagement, evidenced by high affective, behavioral, and cognitive engagement levels. The study identifies three effective strategies teachers use: personalized feedback, gamification, and interactive writing assignments. The research findings show the potential benefits of integrating AI chatbots into EFL writing instruction, facilitating informed decisions to optimize technology usage through understanding student engagement levels and effective teaching strategies, eventually enhancing student learning outcomes.

No abstract available

Context-aware AR instruction enables adaptive and in-situ learning experiences. However, hardware limitations and expertise requirements constrain the creation of such instructions. With recent developments in Generative Artificial Intelligence (Gen-AI), current research tries to tackle these constraints by deploying AI-generated content (AIGC) in AR applications. However, our preliminary study with six AR practitioners revealed that the current AIGC lacks contextual information to adapt to varying application scenarios and is therefore limited in authoring. To utilize the strong generative power of GenAI to ease the authoring of AR instruction while capturing the context, we developed CARING-AI, an AR system to author context-aware humanoid-avatar-based instructions with GenAI. By navigating in the environment, users naturally provide contextual information to generate humanoid-avatar animation as AR instructions that blend in the context spatially and temporally. We showcased three application scenarios of CARING-AI: Asynchronous Instructions, Remote Instructions, and Ad Hoc Instructions based on a design space of AIGC in AR Instructions. With two user studies (N=12), we assessed the system usability of CARING-AI and demonstrated the easiness and effectiveness of authoring with Gen-AI.

The advancing power and capabilities of artificial intelligence (AI) have expanded the roles of AI in education and have created the possibility for teachers to collaborate with AI in classroom instruction. However, the potential types of teacher-AI collaboration (TAC) in classroom instruction and the benefits and challenges of implementing TAC are still elusive. This study, therefore, aimed to explore different types of TAC and the potential benefits and obstacles of TAC through Focus Group Interviews with 30 Chinese teachers. The study found that teachers anticipated six types of TAC, which are thematized as One Teach, One Observe; One Teach, One Assist; Co-teaching in Stations; Parallel Teaching in Online and Offline Classes; Differentiated Teaching; and Team Teaching. While teachers highlighted that TAC could support them in instructional design, teaching delivery, teacher professional development, and lowering grading load, they perceived a lack of explicit and consistent curriculum guidance, the dominance of commercial AI in schools, the absence of clear ethical guidelines, and teachers' negative attitude toward AI as obstacles to TAC. These findings enhance our understanding of how TAC could be structured at school levels and direct the implications for future development and practice to support TAC.

AI technologies transform language instruction by offering feedback, support, and guidance to students, ultimately leading to a more effective and efficient learning experience. The present study investigated the impacts of integrating Writerly and Google Docs to enhance EFL writing instruction. It also assessed students’ perceptions towards using these AI technologies. The study employed a quasiexperimental pretest-posttest two-group design. It used a mixed-methods approach, utilizing tests, questionnaires, focus group discussions, and teacher diaries to gather data from a sample of 92 randomly selected participants. In the experimental group, students enhanced their writing skills through the integration of Writerly and Google Docs, while the control group students received instruction using the traditional paper and pencil feedback system. When the quantitative data were analyzed through independent samples T-test and descriptive statistics, the qualitative data were analyzed thematically. The results confirmed that the integration of Writerly and Google Docs AI technologies, significantly improved EFL writing instruction, as evidenced by a statistically significant difference in writing performance between the experimental and control groups. Hence, students who learned through the integration of Writerly and Google Docs showed improved writing performance as they were able to produce essays that effectively addressed task achievement, coherence and cohesion, lexical resource, and grammatical range and accuracy, whereas those who learned through the conventional method were less effective in producing quality essays. The findings also revealed that the experimental group students had positive perceptions towards integrating Writerly and Google Docs because they found these AI writing technologies interesting, effective, goal-oriented, and supportive. Consequently, this study recommends researchers, curriculum designers, material designers, teachers, and students pay due attention to Writerly and Google Docs.

ABSTRACT In this paper, the author demonstrates their experiences using generative AI to both assist in developing class activity ideas and in facilitating appropriate student use of generative AI in an information literacy course. Attention is given to emphasizing improper uses of generative AI, specifically within the research process, and how the tools may instead be used in an ethical and useful manner to assist with brainstorming research topics. The Association of College & Research Libraries’ Framework for Information Literacy for Higher Education is consulted and used as a guideline for identifying how to best incorporate emerging generative AI tools into information literacy instruction. ChatGPT, Gemini, and Copilot are the three generative AI models that are highlighted and compared, both by the instructor and by students in the author’s information literacy course. The author describes the activities in detail, including how generative AI was used to assist in forming ideas for an interactive lesson to demonstrate various applications of the technology. The process of using generative AI to augment activity planning, classroom experience in implementation, and future considerations are all discussed.

This study explores the integration of Generative AI technology for cognitive offloading in the context of English essay writing, with the aim of promoting inclusive and equitable quality education. Focusing on how AI-assisted cognitive offload instruction enhances students' critical thinking skills, the research adopts a quantitative approach, collecting data from 240 first-year English majors across four colleges through purposive sampling. Statistical analysis reveals that the implementation of Generative AI-enabled Cognitive Offload instruction significantly improves students' critical thinking skills in essay writing. This instructional strategies and specific practices fosters the development of analytical thinking, problem-solving, and effective communication, creating an educational environment that supports originality, critical thought, and responsible writing practices. The findings contribute to the growing body of literature on the role of AI in educational settings, highlighting the potential of AI-assisted cognitive offloading to advance critical thinking and writing skills through structured instructional design.

With the increasing integration of artificial intelligence into education, traditional instructional models in Hydraulic Engineering are shifting toward competence- and performance-oriented pedagogy under the New Engineering framework. Rooted in constructivist and learner-centered theories, this study examines how AI-assisted versus traditional instruction influences learning behaviors, competence development, and academic achievement in engineering education through a quasi-experimental study involving 102 undergraduate students. Results indicate that while the AI-assisted group achieved significantly higher Midterm Report Scores and PPT Presentation Scores, no significant difference was observed in Final Exam Scores between the two groups. Multivariate regression and latent profile analysis reveal that AI-assisted instruction enhances Classroom Participation, Data Processing Ability, and Comprehensive Analytical Ability, yet falls short in fostering Practical Problem-solving Ability compared to traditional instruction. Path analysis further indicates that AI-assisted instruction improves Academic Achievement indirectly by promoting Learning Behaviors, which in turn foster Competence Development, ultimately contributing to improved Academic Achievement. By addressing a critical gap in the literature on the mechanisms of AI integration in engineering education, this study underscores the importance of optimizing learning processes rather than merely pursuing outcome enhancement, offering theoretical and practical insights for AI-integrated instructional reform in the context of New Engineering education.

The integration of artificial intelligence (AI) in language teaching has emerged as a transformative approach, particularly in the realms of English as a Second Language (ESL) and Chinese as a Foreign Language (CFL). This article explores the potential of AI chatbots as effective tools for enhancing language acquisition. By examining the current landscape of AI in language education, we identify the unique benefits that chatbots bring to the learning process, including personalized interaction, immediate feedback, and continuous engagement. The article delves into the design and implementation of AI chatbot systems tailored for ESL and CFL contexts, highlighting their role in vocabulary development, grammar practice, and conversational skills. Furthermore, it addresses the challenges and limitations of using chatbots in language teaching, proposing strategies for overcoming these obstacles. Through case studies and empirical data, the article demonstrates how AI chatbots can be harnessed to create a dynamic and interactive learning environment that caters to the diverse needs of language learners. Ultimately, this work advocates for the thoughtful integration of AI chatbots to complement traditional teaching methods, thereby paving the way for more effective and accessible language education

The deep integration of Artificial Intelligence (AI) is gradually becoming a key force in innovating the teaching of English as a Foreign Language (EFL). This study aims to assess the practical effects of AI technology in providing customized instructional support and learning pathways in EFL instruction. The study reveals the benefits of AI in the instruction of English vocabulary, utilizing the Apriori algorithm from association rule mining and empirical analysis from survey data of 110 second-year university students across four different majors using AI-powered language learning platforms and AI-powered mobile language learning applications (such as UNIPUS AIGC platform and iTEST, intelligent assessment mobile application). It also deduces related teaching strategies and learning models. The results indicate that the use of AI-powered language learning platforms positively impacts English vocabulary learning outcomes in EFL instruction, and the combined use of AI-powered mobile language learning applications for self-testing and in-class tests effectively enhances vocabulary learning efficiency. The findings and conclusions of this study provide valuable insights for EFL educational practice and demonstrate the potential of AI in boosting the effectiveness of language learning, offering empirical support and guidance for future educational decision-making.

In the context of globalization, adapting to modern educational needs and adopting innovative teaching methods have become increasingly crucial, particularly in the field of children’s aesthetic education. This study explores the integration of scaffolding instruction and AI-driven diffusion models in children’s aesthetic education, with a special focus on teaching the traditional Chinese cultural concept of the Twenty-Four Solar Terms. The study develops a specialized dataset for traditional Chinese paintings of the Twenty-Four Solar Terms and introduces a novel compound loss function to optimize the AI models’ training process, thus enhancing the quality of instructional image resources. A scaffolding teaching framework, supported by an AI-driven diffusion model, is established to provide systematic and structured guidance tailored to children’s learning needs. The experimental results indicate that the proposed approach significantly enhances students’ engagement and comprehension of traditional cultural concepts. Specifically, students demonstrated a deeper understanding of the symbolic and artistic meanings embedded in the Twenty-Four Solar Terms, which leads to enhanced cultural appreciation and critical thinking. Moreover, the approach fostered active participation in learning activities, with students exhibiting increased interaction with the educational content. These improvements were particularly evident in the way students creatively interpreted cultural symbols and applied these concepts in their own artistic expressions. This study confirms the potential of AI-driven diffusion models to support more effective teaching practices in aesthetic education, offering valuable insights for integrating modern technology with traditional cultural education and providing key theoretical and practical references for future reforms in children’s aesthetic education. These findings possess significant practical implications, particularly within the education domain. The proposed AI-driven scaffolding teaching method can be broadly applied to classroom instruction in traditional Chinese painting, and it can also be extended to other domains, namely cultural and art education. By generating high-quality instructional image resources, the approach empowers educators to implement personalized teaching strategies in classrooms with diverse cultural backgrounds, while simultaneously enhancing students’ cultural understanding and creativity. Furthermore, for distance and online education, the method possesses a potentially broad application scope, equipping educators with an effective tool for facilitating educational reform and innovation.

This research evaluates the impact of Cami AI integration across SAMR stages (Cami AI-SAMR) in EFL writing instruction. By examining student achievement and perceptions, it explores how AI technology redefines language learning and teaching in diverse educational settings. Through a mixed-method approach with an explanatory sequential research design, this study investigates the quantitative effects of Cami AI-SAMR implementation on student performance and gauges the qualitative responses of 126 EFL university students to its effectiveness and perceptions. The findings show that Cami AI-SAMR implementation impacted significantly EFL students’ writing achievement. Then, the majority of students also had positive perceptions due to the Cami AI’s efficacy in supporting EFL writing learning. These findings provide valuable insights into the transformative potential of Cami AI technology in enhancing EFL pedagogy through the SAMR framework, addressing the diverse needs of students, and reshaping the language education landscape. This research contributes to the ongoing discourse on AI integration in education and offers recommendations for optimizing AI-powered EFL instruction for better learning outcomes and experiences.

Generative AI presents significant opportunities for instructional designers to revolutionize content creation and personalization in online learning environments. This paper explores how generative AI can streamline content generation processes, enhance adaptability to individual learner needs, and improve feedback mechanisms, ultimately fostering more engaging and inclusive learning experiences. Alongside its benefits, generative AI also poses ethical considerations and potential risks, such as perpetuating biases or disrupting the learning process. Navigating these complexities requires a deliberate approach to design deliberation, that involves careful analysis, discussion, and decision-making throughout the design process. This paper proposes a conceptual framework to support instructional designers in leveraging generative AI to promote inclusivity within their design deliberations, emphasizing the importance of addressing ethical considerations and engaging in iterative design practices.

The options for Artificial intelligence (AI) tools used in teacher education are increasing daily, but more is only sometimes better for teachers working in already complex classroom settings. This team discusses the increase of AI in schools and provides an example from administrators, teacher educators, and computer scientists of an AI virtual agent and the research to support student learning and teachers in classroom settings. The authors discuss the creation and potential of virtual characters in elementary classrooms, combined with biometrics and facial emotional recognition, which in this study has impacted student learning and offered support to the teacher. The researchers share the development of the AI agent, the lessons learned, the integration of biometrics and facial tracking, and how teachers use this emerging form of AI both in classroom-based center activities and to support students’ emotional regulation. The authors conclude by describing the application of this type of support in teacher preparation programs and a vision of the future of using AI agents in instruction.

Sung, Min-Chang and Sooyeon Kang. 2024. Developing AI chatbots for pragmatic instruction of Korean secondary L2 English learners. Korean Journal of English Language and Linguistics 24, 441-459. This study explores the design and application of AI chatbots tailored for L2 English pragmatic instruction. Addressing the importance of pragmatic competence alongside the challenges faced by L2 learners, we developed four chatbots through Dialogflow CX and integrated them onto web user interfaces to examine the effectiveness of the AI chatbots in L2 pragmatic learning. We invited a cohort of six middle school students to practice the conversations in pragmatically intensive (i.e., PDR-high) situations. Our findings reveal the need for a dual structure in chatbot design, a separation of technological operation from dialogic content, and a pair-wise page design for cohesive conversations. Additionally, user interfaces should offer contextual clues and linguistic supports to assist learners in understanding and navigating pragmatic exchanges. The participants’ positive perceptions highlight the chatbots’ effectiveness in improving pragmatic awareness and knowledge, akin to other dialogue-based computer-assisted language learning systems. However, the need for guiding feedback mechanisms and inclusive training data for chatbot development has also been noted. In conclusion, chatbots show promise for L2 pragmatic instruction especially when their design architectures and user interfaces carefully reflect dialogue contents, information technology, and supportive elements for effective learner interaction.

No abstract available

Background In 2020, school closures during the COVID-19 pandemic forced students all over the world to promptly alter their learning routines from in-person to distance learning. However, so far, only a limited number of studies from a few countries investigated whether school closures affected students’ performance within intelligent tutoring system—such as intelligent tutoring systems. Method In this study, we investigated the effect of school closures in Austria by evaluating data (n = 168 students) derived from an intelligent tutoring system for learning mathematics, which students used before and during the first period of school closures. Results We found that students’ performance increased in mathematics in the intelligent tutoring system during the period of school closures compared to the same period in previous years. Conclusion Our results indicate that intelligent tutoring systems were a valuable tool for continuing education and maintaining student learning during school closures in Austria.

Programming skills are rapidly becoming essential for many educational paths and career opportunities. Yet, for many international students, the traditional approach to teaching introductory programming courses can be a significant challenge due to the complexities of the language, the lack of prior programming knowledge, and the language and cultural barriers. This study explores how large language models and gamification can scaffold coding learning and increase Chinese students’ sense of belonging in introductory programming courses. In this project, a gamification intelligent tutoring system was developed to adapt to Chinese international students’ learning needs and provides scaffolding to support their success in introductory computer programming courses. My research includes three studies: a formative study, a user study of an initial prototype, and a computer simulation study with a user study in progress. Both qualitative and quantitative data were collected through surveys, observations, focus group discussions and computer simulation. The preliminary findings suggest that GPT-3-enhanced gamification has great potential in scaffolding introductory programming learning by providing adaptive and personalised feedback, increasing students’ sense of belonging, and reducing their anxiety about learning programming.

Self-regulated learning (SRL) is critical for learning across tasks, domains, and contexts. Despite its importance, research shows that not all learners are equally skilled at accurately and dynamically monitoring and regulating their self-regulatory processes. Therefore, learning technologies, such as intelligent tutoring systems (ITSs), have been designed to measure and foster SRL. This paper presents an overview of over 10 years of research on SRL with MetaTutor, a hypermedia-based ITS designed to scaffold college students’ SRL while they learn about the human circulatory system. MetaTutor’s architecture and instructional features are designed based on models of SRL, empirical evidence on human and computerized tutoring principles of multimedia learning, Artificial Intelligence (AI) in educational systems for metacognition and SRL, and research on SRL from our team and that of other researchers. We present MetaTutor followed by a synthesis of key research findings on the effectiveness of various versions of the system (e.g., adaptive scaffolding vs. no scaffolding of self-regulatory behavior) on learning outcomes. First, we focus on findings from self-reports, learning outcomes, and multimodal data (e.g., log files, eye tracking, facial expressions of emotion, screen recordings) and their contributions to our understanding of SRL with an ITS. Second, we elaborate on the role of embedded pedagogical agents (PAs) as external regulators designed to scaffold learners’ cognitive and metacognitive SRL strategy use. Third, we highlight and elaborate on the contributions of multimodal data in measuring and understanding the role of cognitive, affective, metacognitive, and motivational (CAMM) processes. Additionally, we unpack some of the challenges these data pose for designing real-time instructional interventions that scaffold SRL. Fourth, we present existing theoretical, methodological, and analytical challenges and briefly discuss lessons learned and open challenges.

In the realm of online tutoring intelligent systems, e-learners are exposed to a substantial volume of learning content. The extraction and organization of exercises and skills hold significant importance in establishing clear learning objectives and providing appropriate exercise recommendations. Presently, knowledge graph-based recommendation algorithms have garnered considerable attention among researchers. However, these algorithms solely consider knowledge graphs with single relationships and do not effectively model exercise-rich features, such as exercise representativeness and informativeness. Consequently, this paper proposes a framework, namely, the Knowledge Graph Importance-Exercise Representativeness and Informativeness Framework, to address these two issues. The framework consists of four intricate components and a novel cognitive diagnosis model called the Neural Attentive Cognitive Diagnosis model to recommend the proper exercises. These components encompass the informativeness component, exercise representation component, knowledge importance component, and exercise representativeness component. The informativeness component evaluates the informational value of each exercise and identifies the candidate exercise set E C that exhibits the highest exercise informativeness. Moreover, the exercise representation component utilizes a graph neural network to process student records. The output of the graph neural network serves as the input for exercise-level attention and skill-level attention, ultimately generating exercise embeddings and skill embeddings. Furthermore, the skill embeddings are employed as input for the knowledge importance component. This component transforms a one-dimensional knowledge graph into a multidimensional one through four class relations and calculates skill importance weights based on novelty and popularity. Subsequently, the exercise representativeness component incorporates exercise weight knowledge coverage to select exercises from the candidate exercise set for the tested exercise set. Lastly, the cognitive diagnosis model leverages exercise representation and skill importance weights to predict student performance on the test set and estimate their knowledge state. To evaluate the effectiveness of our selection strategy, extensive experiments were conducted on two types of publicly available educational datasets. The experimental results demonstrate that our framework can recommend appropriate exercises to students, leading to improved student performance.

Abstract Learning through an Intelligent Tutoring System (ITS) lies in performing well in academics and improving the learner’s learning outcomes. The recent developments within an Intelligent Tutoring System are not focused on improving human–computer interactions. The best way to overcome this constraint is to develop ITS interfaces to provide learners with better learning experiences. In this study, Augmented Reality (AR) potential along with AI methodologies is utilized within the developed ITS interface to improve the learning experience of the learning-disabled learners. Augmented Reality is where virtual images overlay the physical world, and Mixed Reality is an emerging technology that presents an altered reality, which engages users to interact with an environment developed using virtual objects. With several tools and applications available, creating and providing immersive learning experiences for learners is rising. AR in educating the specially-abled is widespread, and its benefits are sufficiently explored. However, limited AR applications are designed specifically for supporting the education of individuals with learning disabilities. The available application, and their impact on learning-disabled learners, needs detailed investigation. This work presents an Intelligent Tutoring System (ITS) to educate learners with Augmented Reality (AR) based content. The ITS learner module implemented in this study was developed for learning disabilities identification and we have assessed total 105 participants (with or without Learning Disabilities) for the experiment. The ITS performance is compared (with or without) AR content-based learning and based on the findings, AR-based learning through ITS is effective. The benefits are manifold, increase in motivation, ease of interaction, development in cognitive skills, enhancement in short-term memory, and making lessons more enjoyable, turning the overall experience stimulating and engaging.

Nowadays, the improvement of digital learning with Artificial Intelligence has attracted a lot of research, as it provides solutions for individualized education styles which are independent of place and time. This is particularly the case for computer science, as a tutoring domain, which is rapidly growing and changing and as such, learners need frequent update courses. In this paper, we present a thorough evaluation of a fuzzy-based intelligent tutoring system (ITS), that teaches computer programming. The evaluation concerns multiple aspects of the ITS. The evaluation criteria are: (i) context, (ii) effectiveness, (iii) efficiency, (iv) accuracy, (v) usability and satisfaction, and (vi) engagement and motivation. In the evaluation process students of an undergraduate program in Informatics of the University of Piraeus in Greece participated. The evaluation method that was used included questionnaires, analysis of log files and experiments. Also, t-tests were conducted to certify the validity of the evaluation results. Indeed, the evaluation results are very positive and show that the incorporated fuzzy mechanism to the presented ITS enhances the system with Artificial Intelligence and through this, it increases the learners’ satisfaction and new knowledge learning and mastering, improves the recommendation accuracy of the system, the efficacy of interactions, and contributes positively to the learners’ engagement in the learning process.

Intelligent Tutoring Systems (ITSs) are educational systems that reflect knowledge using artificial intelligence implements. In this paper, we give an outline of the Programming-Tutor architectural design with the core implements on user interaction. This pilot proposal is for designing a model domain of a subset in the computer programming language. The completed project would be adequate to show the idea of a completely developed computing Intelligent Tutoring System in online programming courses to offer benefits to students in the Pacific. This proposed concept would also provide students with an immersive learning experience in an online course to assist in a formative assessment to enhance student learning. A smart tutoring system can provide prompt input of high quality which not only conveys to students about the consistency of the solution but also provides them with information on the precision of the key concerning their existing solutions expertise. This Intelligent Tutoring System (ITS) is proposed to be designed using intelligent algorithms such as optimized ant colony to be able to support the online tutoring system that can initiate the complex learning principles in computing science courses. It is also hypothesized that, based on the performance of other Intelligent Tutoring Systems, students would be able to learn to program more easily in regional campuses and acquire experiences more rapidly and efficiently than students who are taught using conventional methods in an online mode.

Background: Skill integration is vital in students' mastery development and is especially prominent in developing code tracing skills which are foundational to programming, an increasingly important area in the current STEM education. However, instructional design to support skill integration in learning technologies has been limited. Objectives: The current work presents the development and empirical evaluation of instructional design targeting students' difficulties in code tracing particularly in integrating component skills in the Trace Table Tutor (T3), an intelligent tutoring system. Methods: Beyond the instructional features of active learning, step-level support, and individualized problem selection of intelligent tutoring systems (ITS), the instructional design of T3 (e.g., hints, problem types, problem selection) was optimized to target skill integration based on a domain model where integrative skills were represented as combinations of component skills. We conducted an experimental study in a university-level introductory Python programming course and obtained three findings. Results and Conclusions: First, the instructional features of the ITS technology support effective learning of code tracing, as evidenced by significant learning gains (medium-to-large effect sizes). Second, performance data supports the existence of integrative skills beyond component skills. Third, an instructional design focused on integrative skills yields learning benefits beyond a design without such focus, such as improving performance efficiency (medium-to-large effect sizes). Major Takeaways: Our work demonstrates the value of designing for skill

No abstract available

No abstract available

We propose and implement a novel intelligent tutoring system, called RadarMath, to support intelligent and personalized learning for math education. The system provides the services including automatic grading and personalized learning guidance. Specifically, two automatic grading models are designed to accomplish the tasks for scoring the text-answer and formula-answer questions respectively. An education-oriented knowledge graph with the individual learner’s knowledge state is used as the key tool for guiding the personalized learning process. The system demonstrates how the relevant AI techniques could be applied in today's intelligent tutoring systems.

No abstract available

A common goal of Intelligent Tutoring Systems (ITS) is to provide learning environments that adapt to the varying abilities and characteristics of users. This type of adaptivity is possible only if the ITS has information that characterizes the learning behaviors of its users and can adjust its pedagogy accordingly. This study investigated an intelligent tutoring system with computer agents (AutoTutor) designed to improve comprehension skills in adults with low reading literacy. One goal of this study was to classify adults into different clusters based on their behavioral patterns (accuracy and response time to answer questions) while they interacted with AutoTutor to help them improve their reading comprehension skills. A second goal was to investigate whether adults’ behaviors were associated with different reading components. A third goal was to assess improvements in reading comprehension skills, based on psychometric tests, of different clusters of readers. Performance on AutoTutor was collected in a targeted 100-hour hybrid intervention for adult readers (n = 252) that included both human teachers and the AutoTutor system. The adults’ average accuracy and response time in AutoTutor were used to cluster the adults into four categories: higher performers (comparatively fast and accurate), conscientious readers (slow but accurate), under-engaged readers (fast at the expense of somewhat lower accuracy) and struggling readers (slow and inaccurate). Two psychometric tests of comprehension were used to assess comprehension. Gains in comprehension scores were highest for conscientious readers, lowest for struggling readers, with higher performing readers and under-engaged readers in between. The results provide guidance to enhance the adaptivity of AutoTutor.

We investigate how automated, data-driven, personalized feedback in a large-scale intelligent tutoring system (ITS) improves student learning outcomes. We propose a machine learning approach to generate personalized feedback, which takes individual needs of students into account. We utilize state-of-the-art machine learning and natural language processing techniques to provide the students with personalized hints, Wikipedia-based explanations, and mathematical hints. Our model is used in Korbit (https://www.korbit.ai), a large-scale dialogue-based ITS with thousands of students launched in 2019, and we demonstrate that the personalized feedback leads to considerable improvement in student learning outcomes and in the subjective evaluation of the feedback.

The way in which people learn and institutions teach is changing due to the ever-increasing impact of technology. People access the Internet anywhere, anytime and request online training. This has brought about the creation of numerous online learning platforms which offer comprehensive and effective educational solutions which are 100% online. These platforms benefit from intelligent tutoring systems that help and guide students through the learning process, emulating the behavior of a human tutor. However, these systems give the student little freedom to experiment with the knowledge of the subject, that is, they do not allow him/her to propose and carry out tasks on his/her own initiative. They are very restricted systems in term of what the student can do, as the tasks are defined in advance. An intelligent tutoring system is proposed in this paper to encourage students to learn through experimentation, proposing tasks on their own initiative, which involves putting into use all the skills, abilities tools and knowledge needed to successfully solve them. This system has been designed developed and applied for learning predictive parsing techniques and has been used by Computer Science students during four academic courses to evaluate its suitability for improving the student’s learning process.

Sketching is a practical and useful skill that can benefit communication and problem solving. However, it remains a difficult skill to learn because of low confidence and motivation among students and limited availability for instruction and personalized feedback among teachers. There is an need to improve the educational experience for both groups, and we hypothesized that integrating technology could provide a variety of benefits. We designed and developed an intelligent tutoring system for sketching fundamentals called Sketchtivity, and deployed it in to six existing courses at the high school and university level during the 2017-2018 school year. 268 students used the tool and produced more than 116,000 sketches of basic primitives. We conducted semi-structured interviews with the six teachers who implemented the software, as well as nine students from a course where the tool was used extensively. Using grounded theory, we found ten categories which unveiled the benefits and limitations of integrating an intelligent tutoring system for sketching fundamentals in to existing pedagogy.

Abstract Students nowadays are hard to be motivated to solve logical problems with traditional teaching methods. Computers, Smartphone's, tablets and other smart devices disturb their attention. But those smart devices can be used as auxiliary tools of modern teaching methods. The flipped classroom is one such innovative method that moves the solving problems outside the classroom via technology and reinforces solving problems inside the classroom via learning activities. In this paper, the authors implement flipped classroom as an element of Internet of Things (IOT) into learning process of mathematical logic course. In the flipped classroom, an Intelligent Tutoring System (ITS) was used to help students work with the problems in the course outside the classroom. This study showed that perceived usefulness, self-efficacy, compatibility, and perceived support for enhancing social ties are important antecedents to continuance intention to use flipped classroom.

The idea of clustering students according to their online learning behavior has the potential of providing more adaptive scaffolding by the intelligent tutoring system itself or by a human teacher. With the aim of identifying student groups who would benefit from the same intervention in AC-ware Tutor, this research examined online learning behavior using 8 tracking variables: the total number of content pages seen in the learning process; the total number of concepts; the total online score; the total time spent online; the total number of logins; the stereotype after the initial test, the final stereotype, and the mean stereotype variability. The previous measures were used in a four-step analysis that consisted of data preprocessing, dimensionality reduction, the clustering, and the analysis of a posttest performance on a content proficiency exam. The results were also used to construct the decision tree in order to get a human-readable description of student clusters.

No abstract available

Distance Education Courses in Virtual Learning Environments (VLE) employ tutors for pedagogical support, monitoring of students, and detecting possible dropouts. However, when courses have a higher number of students, there can be a work overload on human tutors, impacting their work quality. To mitigate this problem, artificial smart tutors can be used not only to increase the capacity to meet students' needs but also to improve the monitoring of their performance. Increasing the scalability of online courses can be achieved using an intelligent artificial tutor. Goal: Analyze the viability of an intelligent tutor in meeting the main tutoring demands made by students on Dell Accessible Learning platform. Method: We developed an artifact called STUART that monitors DAL and interacts with students, providing automation, intelligence, and support to the teaching and learning process. We programmed STUART to meet reactively and proactively students' main demands, based on the corpus of interactions scenarios on previous courses. Fourteen participants attended two classes of a Distance Education Course, with and without STUART, where interaction data were collected. Results: For 76% of the participants, STUART helped solve the more frequent problems in the pedagogical, technical, and content levels. There was an average reduction of 87% in the requests for a human tutor and a reduction of 27% in the time needed to finish tasks. That resulted in the preference of STUART compared to the human tutor for 85% of the students. SUS (usability assessment) scored 86.

No abstract available

No abstract available

The Office of Naval Research (ONR) organized a STEM Challenge initiative to explore how intelligent tutoring systems (ITSs) can be developed in a reasonable amount of time to help students learn STEM topics. This competitive initiative sponsored four teams that separately developed systems that covered topics in mathematics, electronics, and dynamical systems. After the teams shared their progress at the conclusion of an 18-month period, the ONR decided to fund a joint applied project in the Navy that integrated those systems on the subject matter of electronic circuits. The University of Memphis took the lead in integrating these systems in an intelligent tutoring system called ElectronixTutor. This article describes the architecture of ElectronixTutor, the learning resources that feed into it, and the empirical findings that support the effectiveness of its constituent ITS learning resources. A fully integrated ElectronixTutor was developed that included several intelligent learning resources (AutoTutor, Dragoon, LearnForm, ASSISTments, BEETLE-II) as well as texts and videos. The architecture includes a student model that has (a) a common set of knowledge components on electronic circuits to which individual learning resources contribute and (b) a record of student performance on the knowledge components as well as a set of cognitive and non-cognitive attributes. There is a recommender system that uses the student model to guide the student on a small set of sensible next steps in their training. The individual components of ElectronixTutor have shown learning gains in previous decades of research. The ElectronixTutor system successfully combines multiple empirically based components into one system to teach a STEM topic (electronics) to students. A prototype of this intelligent tutoring system has been developed and is currently being tested. ElectronixTutor is unique in its assembling a group of well-tested intelligent tutoring systems into a single integrated learning environment.

No abstract available

No abstract available

There are significant challenges involved in the design and implementation of a dialog-based tutoring system (DBT) ranging from domain engineering to natural language classification and eventually instantiating an adaptive, personalized dialog strategy. These issues are magnified when implementing such a system at scale and across domains. In this paper, we describe and reflect on the design, methods, decisions and assessments that led to the successful deployment of our AI driven DBT currently being used by several hundreds of college level students for practice and self-regulated study in diverse subjects like Sociology, Communications, and American Government.

No abstract available

Nowadays, society is in constant evolution, which allows constant production of new knowledge. In this way, citizens are constantly pressured to obtain new qualifications through training/requalification. The need for qualified people has been growing exponentially, which means that resources for education/training are limited to being used more efficiently. In this paper we will focus in the design the user model, so, we propose an innovative approach to design a user model that monitors the user’s biometric behaviour by measuring their level of attention during e-learning activities. In addition, a machine learning categorization model is presented that oversees user activity during the session. We intend to use non-invasive methods of intelligent tutoring systems, observing the interaction of users during the session. Furthermore, this article highlights the main biometric behavioural variations for each activity and bases the set of attributes relevant to the development of machine learning classifiers to predict users’ learning preference. The results show that there are still mechanisms that can be explored and improved to better understand the complex relationship between human behaviour, attention and evaluation that could be used to implement better learning strategies. These results can be decisive in improving ITS in e-learning environments and to predict user behaviour based on their interaction with technology devices.

No abstract available

No abstract available

According to the educational regulations in Taiwan, students are required to learn English when they are at the first grade of elementary school. However, not all the students have an appropriate environment to practice English, especially, for those students whose school is not located in the city. Thus, their English abilities in speaking, reading, and listening are poor. An intelligent tutoring system is used to help the students improve their English capabilities. This paper aims to provide a convenient tutoring environment, where teachers and students do not need to prepare a lot of teaching aids. They can teach and learn English whenever in the environment. Also, it proposes a method to verify the intelligent tutoring system using Petri nets. We have built the intelligent tutoring system based on Augmented Reality (AR), Text-to-Speech (TTS), and Speech Recognition (SR). This intelligent tutoring system is divided into two parts: one for teachers and the other for students. The experimental results have indicated that using Petri nets can help users verify the intelligent tutoring system for better learning performance and operate it correctly.

No abstract available

This full paper of research to practice category describes an experiment using an Intelligent Tutoring System to support teaching and learning that evolve solutions of first-degree equations. The difficulties encountered in solving algebra problems can make students consider math and related science a difficult task. We propose a Tutor to promote the learning of this class of problems. In this article, we describe the functioning and architecture of the tutor composed by the Interface, Student Model, Tutorial Strategies and Domain Model. Also, a Virtual Assistant was included to motivate students in the learning process. The tutoring strategies that comprise the tutor are projected from models found in the literature. To approximate the interactions of the tutor with the actions of the teachers, Lev Vygotsky’s Concept of Zone Proximal Development was used. The methodology used in this project was the case study applied in a real-life context, obtaining a combination of quantitative and qualitative evidence about technical and pedagogical usability that show the degree of satisfaction of the students regarding the proposed approach.

No abstract available

A comparison of machine learning models developed for an adaptive learning platform is done, aiming to revolutionize educational methodologies in response to the diverse learning needs of individuals. The model dynamically adjusts the difficulty of test questions based on user performance, creating a personalized learning experience. The study evaluates and compares the model's performance across various machine learning algorithms, including Random Forest, Decision Tree, Naive Bayes and Logistic Regression. The findings reveal the model's high accuracy, ranging from 66% to 92%, showcasing its efficacy in adapting to individual capabilities. By generating user performance profiles, the model successfully tailors challenges to meet diverse learning needs, contributing to an optimized educational experience. Moreover, this research contributes to the field by demonstrating that the proposed model's simplicity, particularly when using the Decision Tree algorithm, yields comparable accuracy results to more complex models. Through extensive testing, the model consistently performed with accuracy ranging from 66% to 92%, even with a small preliminary test size of 21 questions. This simplicity, coupled with its ability to dynamically adjust difficulty levels, positions the model as a pragmatic and effective solution for adaptive learning. The study concludes by underlining the model's effectiveness in providing personalized learning experiences, ultimately enhancing user engagement and educational outcomes.

Introduction: Educational landscapes have been transformed by technological advancements, enabling adaptive and flexible learning through AI-based and decision-oriented adaptive learning systems. The increasing importance of this solutions is underscored by the pivotal role of the learner model, representing the core of the teaching-learning dynamic. This model, encompassing qualities, knowledge, abilities, behaviors, preferences, and unique distinctions, plays a crucial role in customizing the learning experience. It influences decisions related to learning materials, teaching strategies, and presentation styles. Objective: This study meets the need for applying AI-driven adaptive learning in education, implementing a novel method that uses self-esteem (ES), emotional intelligence (EQ), and demographic data to predict student performance and adjust the learning process. Methods: Our study involved collecting and processing data, constructing a predictive machine learning model, implementing it as an online solution, and conducting an experimental study with 146 high school students in computer science and French as foreign language. The aim was to tailor the teaching-learning process to the learners' performance. Results: Significant correlations were observed between self-esteem, emotional intelligence, demographic data, and final grades. The predictive model demonstrated a 90% accuracy rate. In the experimental group, the results indicated higher scores, with an average of 15.78/20 compared to the control group's 12.53/20 in computer science. Similarly, in French as a foreign language, the experimental group achieved an average of 13.78/20, surpassing the control group's 10.47/20. Conclusion: The achieved results motivate the creation of a multifactorial AI-driven adaptive learning platform. Recognizing the necessity for improvement, we aim to refine the predicted performance score through the incorporation of a diagnostic evaluation, ensuring an optimal grouping of learners.

In the current post-pandemic context, learning mediated by virtual platforms has become the de facto methodology. Despite its wide use and recent developments, virtual learning still faces several challenges to reach its maximal potential. One of these challenges is related to the homogeneity of the contents presented to the students and the lack of awareness of the different learning styles. This scenario has evidenced the need to develop mechanisms that strengthen learning processes according to the particular preferences of each student. In this context, this work presents a methodology to implement a didactic strategy oriented to incorporate adaptive learning in virtual environments. The strategy is supported by a machine learning based contents recommender that is integrated with the Learning Management System to improve the student's outcomes.

No abstract available

Recent advancements in artificial intelligence have paved the way for innovative personalized education through Interactive Machine Learning Systems (IMLS). This study explores how combining large language models (LLMs) with knowledge graphs (KGs) within IMLS can create adaptive, customized learning experiences. Leveraging the sophisticated natural language understanding and generation capabilities of LLMs, the system enables dynamic interactions between learners and educational content, offering real-time feedback and personalized guidance. At the same time, knowledge graphs provide a well-organized framework for structuring educational materials, detailing the relationships among concepts, skills, and learner profiles. By integrating these technologies, the system can intelligently assess individual learner needs, predict learning trajectories, and tailor instructional strategies accordingly. Empirical evaluations across various educational settings have shown significant enhancements in learner engagement, knowledge retention, and overall academic performance. This research highlights the promising potential of merging advanced language models with semantic knowledge representations to develop scalable and effective personalized education solutions that cater to the unique needs of every learner. This study addresses the existing gap in combining large language models and knowledge graphs for personalized education by proposing a novel integrated framework. Our empirical results demonstrate significant improvements in learner engagement and knowledge retention, highlighting the robustness and scalability of the approach.

This research investigates the development of an adaptive blended teaching model (ABTM) that employs customized instructional strategies to optimize learning outcomes. The approach harnesses data-driven insights derived from student performance, behavior, and engagement to provide a personalized educational experience tailored to each student’s requirements. The integration of big data analytics and machine learning (ML) in education presents significant potential to transform traditional teaching methodologies. Data is collected from open sources, including engagement scores, assessment results, forum participation, attendance, and study hours. Preprocessing steps include data cleaning, normalization, and handling missing values to ensure data reliability. The term frequency-inverse document frequency (TF-IDF) text mining technique is utilized to extract features from student-generated content, highlighting essential phrases. TF-IDF enables the identification of critical learning themes and areas requiring additional support. A hybrid method, namely, the snow leopard optimized-tuned intelligent CatBoost (SLO-ICatBoost), is deployed for predicting student grades, assessing performance, and enhancing the educational process. The SLO improves the selection of relevant features, while the ICatBoost algorithm classifies students based on their performance patterns and learning behaviors. When compared to conventional teaching techniques, the proposed SLO-ICatBoost method significantly improves precision (0.990), accuracy (0.991), F1-score (0.990), and recall (0.991). Due to its flexibility in accommodating various learning environments and individual requirements, this approach can be applied in diverse educational settings.

This paper investigates how ML techniques could be applied to design and enhance AITS with the aim of offering personalized, data-driven learning environments.   Growing need for smart educational tools motivates innovative ML integration into tutoring systems to meet different student demands, increase participation, and improve academic performance.   The study evaluates the efficacy of significant ML algorithms—including supervised learning, RL, and DL—within the context of learner modeling, adaptive content delivery, and real-time feedback systems.   A prototype ML-AITS framework was developed and tested across multiple learner groups, comparing traditional education, basic adaptive systems, and fully adaptable ML-based systems.   Quantitative research reveals that ML-AITS substantially surpasses traditional methods in key areas such learner participation, instructional effectiveness, and learning results.   For instance, pupils using ML-AITS exhibited up to a 16.9% rise in post-test scores and more active measures compared to their counterparts.   Comprising learner profile, adaptive content delivery, real-time assessment, performance analytics, and continuous learning layers, the proposed five-layered ML-AITS architecture forms a dynamic and intelligent ecosystem competent of self-improvement.   The findings validate the potential of machine learning to change digital education by means of intelligent personalisation and adaptive feedback loops.   Our work contributes to the growing field of educational technology by providing a scalable and efficient ML-driven tutoring system.   It also offers a foundation for future studies on more general applications in multilingual and multicultural educational environments, emotionally intelligent systems, and NLP integration.

Although technology has significantly improved the teaching and learning process, it has not been able to increase students' self-motivation and engagement at the same level. The lack of self-motivation and intermittent engagement is currently one of the primary challenges faced by educators. This new approach to learning called the hybrid gamification framework uses a combination of artificial intelligence (AI), machine learning (ML), and the Adaptive Neuro-Fuzzy Inference System (ANFIS) to create a more engaging and personalized learning experience. By tracking students' interactions and performance, the system can allocate rewards based on their progress, which helps to increase their motivation and engagement. This technology makes it possible for educators to collect and analyse data related to students' engagement patterns, quiz scores, time spent on learning activities, participation in discussion forums, and much more. This data analysis enables educators to identify struggling students and high achievers, allowing them to provide tailored support and instruction to maximize student success. A pilot implementation of this system involving 200 computer science students successfully demonstrated the effectiveness of this technology. This research provides a comprehensive understanding of gamification's impact by combining quantitative data with qualitative insights.

This study explores the integration of supervised machine learning using decision tree algorithms within the framework of the Structure of Observed Learning Outcomes for diagnosing and customizing learning pathways in middle school mathematics education. Focusing on seventh-grade students' proficiency in numbers and operations, the research employs a large dataset of student responses to develop a real-time adaptive diagnostic tool. The tool classifies students into five proficiency levels-starter, basic, medium, high, and advanced-based on their responses. Initial findings demonstrate an overall classification accuracy of 83%, with further analysis revealing specific strengths and weaknesses across different proficiency levels. This research underscores the potential of supervised machine learning to enhance educational diagnostics and contribute to personalized learning experiences, suggesting that such technology can significantly improve educational outcomes by dynamically adjusting to individual student needs.

This paper explores the application of Artificial Intelligence (AI) and Machine Learning (ML) in adaptive eLearning platforms designed for cybersecurity training, with a focus on entrepreneurs. Due to limited resources and technical expertise, entrepreneurs face unique challenges in protecting their businesses from cyber threats. Traditional training methods often fail to meet their needs, highlighting the importance of AI-driven platforms that offer personalized learning experiences. The paper examines the benefits of AI-powered eLearning systems, including improved engagement, real-time assessments, and adaptation to diverse learning styles. It also addresses emerging trends in AI and ML for cybersecurity education, the integration of adaptive eLearning into entrepreneurial support systems, and the ethical and regulatory implications of AI-driven learning. Finally, recommendations are provided for policymakers and educators to support the growth of AI in cybersecurity training. The findings suggest that AI-powered platforms can offer scalable, effective solutions for entrepreneurs to enhance their cybersecurity skills and protect their digital assets. Keywords: Artificial Intelligence, Machine Learning, Cybersecurity Training, Adaptive eLearning, Entrepreneurs, Personalized Learning.

The aim of this research is to investigate the efficacy of an adaptive educational platform utilizing machine learning in enhancing the quality of higher education, validated through empirical analysis. By assessing individualized learning requirements and behavioral patterns of students, the platform offers tailored learning paths and resource recommendations to enhance learning outcomes and satisfaction. Employing an experimental control group design, the study compares teaching methodologies between the experimental and control groups. Findings indicate significant improvements in academic performance, satisfaction, and learning behavior among students in the experimental group, highlighting the effectiveness of the machine learning-based adaptive educational platform in elevating higher education quality and offering novel avenues for educational enhancement and innovation. This study holds crucial theoretical and practical implications for advancing educational technology and enhancing educational quality.

In an increasingly diversified educational landscape, adaptable curriculum development is critical to meeting the varying requirements of students. Traditional curriculum design approaches frequently lack the flexibility to suit individual student variances, resulting in disengagement and poor learning outcomes. Current approaches are mostly based on static evaluations and generic material delivery, and do not take advantage of the wealth of data accessible regarding student achievement and participation. This work provides a unique way for optimizing adaptive curriculum development using Light Gradient Boosting (LightGBM), a machine learning method known for its effectiveness and predictive capacity. By using real-time data analytics, the suggested system enables personalized learning pathways that dynamically alter content based on student progress and preferences. The overall methodology includes collecting data from several educational sources, pre-processing to guarantee quality, and using LightGBM for predictive modelling. The adaptive curriculum’s efficacy is evaluated using key measures such as pupil involvement, rate of retention, and academic performance. A series of case studies from various educational settings throughout the world are used to evaluate performance, comparing traditional curricula to an adaptive model constructed using LightGBM. Preliminary data show considerable gains in student outcomes, including greater engagement and achievement levels.

Student attrition and academic failure remain pervasive challenges in education, often occurring at substantial rates and posing considerable difficulties for timely identification and intervention. Learning management systems such as Moodle generate extensive datasets reflecting student interactions and enrollment patterns, presenting opportunities for predictive analytics. This study seeks to advance the field of dropout and failure prediction through the application of artificial intelligence with machine learning methodologies. In particular, we employed the CatBoost algorithm, trained on student activity logs from the Moodle platform. To mitigate the challenges posed by a limited and imbalanced dataset, we employed sophisticated data balancing techniques, such as Adaptive Synthetic Sampling, and conducted multi-objective hyperparameter optimization using the Non-dominated Sorting Genetic Algorithm II. We compared models trained on weekly log data against a single model trained on all weeks’ data. The proposed model trained with all weeks’ data demonstrated superior performance, showing significant improvements in F1-scores and recall, particularly for the minority class of at-risk students. For example, the model got an average F1-score across multiple weeks of approximately 0.8 in the holdout test. These findings underscore the potential of targeted machine learning approaches to facilitate early identification of at-risk students, thereby enabling timely interventions and improving educational outcomes.

The solution of a typical programming task involves algorithmic thinking, pattern recognition, decomposition, and abstraction skills. These skills are basic pillars of Computational thinking (CT) and are essential for a computer programmer. Therefore, a programming teacher needs to take these skills into account for the evaluation of students. Existing methods for improving programming competency don’t consider the Computational Thinking of students and perform grading of students without taking these skills into account. Due to this limitation, deficiencies of these skills in students remain unrevealed, posing difficulties for educators to provide need-oriented feedback. The performance of programming students is thus hindered by a lack of interventions. This study proposes a method to evaluate programming students in terms of CT skill components and group them based on their skill set. In this study, 780 students of object-oriented programming (OOP) class were given programming assignments for assessment of programming competencies. These students were then given small programming tasks requiring different computational thinking skill components for their solutions. A machine learning approach was used to perform grouping of these students based on their scores. Six groups of programming students, each having its own set of skill deficiencies, were identified as a result of clustering. One of the groups had an empty set of skill deficiencies and consisted of students proficient in programming. Each of the other five groups had a non-empty set of skill deficiencies and comprised non-proficient programming students. A group-specific skill development plan was built for the groups having skill deficiencies. It was found that such feedback was very effective and improved the CT skill deficiencies of 82% of students.

No abstract available

In order to effectively implement adaptive learning within E-learning systems, it is crucial to accurately define thelearner's profile that reflects the characteristics necessary for optimal learning. Traditional methods of identifying profiles often relyon questionnaires to collect data from learners, which can be time-consuming and result in irrelevant data due to arbitrary responses.As a solution, we propose an intelligent and dynamic model for adaptive learning that takes into account the entire learning process,from diagnostic assessment to knowledge assimilation. Our approach utilizes the k-means classification algorithm to group learners based on similar characteristics, as defined by the KOLB model. To enhance the accuracy of our model, we also incorporate neural networks to automatically predict learning styles and using decision tree to propose a adaptative pedagogical content to learner. By doing so, we aim to improve the overall performance of our proposed model.

Mobile learning is becoming more and more popular today. It gained popularity recently due to the COVID-19 pandemic restrictions in 2020. However, to provide learners with appropriate educational materials in such a mobile environment, the characteristics and context of the learners must be considered. Therefore, in this paper, we propose a framework for providing an adaptive context-aware learning process considering a combination of student learning models and principles of Universal Design for Learning (UDL). The proposed system consists of components capable of detecting changes in context and adapting the way the application responds and behaves. The framework uses a machine-learning algorithm to predict learners’ characteristics and follow UDL principles to deliver enriched user experience and location-aware content and activities. An online survey was conducted with 20 undergraduate students. We analyzed their levels of satisfaction with the proposed m-learning system. From the analyzed data, we noticed that the average rating values are close to 4.5, which indicates that the proposed m-learning system complies with UDL principles and provides an adaptive and localized learning environment, thus enhancing the efficiency of the learning process and experiences. The study also investigated the impact of factors (i.e., noise level, physical activity, and location) on learners’ concentration towards the learning process. The results show that these factors have a significant impact on the learner’s concentration level.

This paper offers AIComprehend, a machine learning-based online platform whose objective is to improve the English reading comprehension of its users using an adaptive learning approach. The project entailed the development of a web-based application that consisted of multiple-choice reading comprehension questions of varying difficulty levels and knowledge components. A four-week beta testing of the application was carried out at Daniel R Aguinaldo National High School where 58 students were divided into control and experimental groups, with the experimental group using the application for 30 minutes daily. The Pre-test and post test were conducted four weeks apart with identical difficulty to assess the intervention. Results showed that the mean Pre-test score of 6.97 for the experimental group improved to a post-test score of 7.94, signifying an approximately 13.9% improvement. The control group, however, saw a decline from 8.46 in the pre-test to 6.19 in the post-test, marking a roughly 26.8% decrease. Moreover, the accuracy (79.06%) and AUC (65.95%) scores of the PFA-based system showed potential as reading comprehension performance tracing tool.

With the increase of the internet, E-learning adaptation and recommnadation system has become a new trend in learning. In contrast to traditional learning (all to all sytem), adaptive e-learning systems provides a special environment in which learners are in control of their learning process. Yet finding the most appropriate learning path and content for a given learner according to his or her profil is a very important issue in achieving the learning objectives. In this context several methods have been proposed with different level of performance in terms of precision and quality of adaptation. In this paper, a new adaptation approach based on learner's motivation level is proposed in E-learning environement combining Content-Based Filtering technique, and machine learning algorithms. The proposed approach, is aimed to built a customized learning path while identifying and selectiong the most appropriate learning objects for learners during the learning activity. Additionally, a few experiments were run to gauge how well our contribution worked. The results obtained by the proposed strategy demonstrate that considering the learner's motivation level enhances and reinforces the level of learning among students during the learning activity.

Within the space of competitive examinations, such as the Union Open Benefit Commission (UPSC) exams, the requirement for personalized and versatile instructive substance proposal frameworks has gotten to be progressively imperative. This inquiry about presents an inventive approach to building a Versatile Instructive Substance Proposal Framework (AECRS) custom-made particularly for UPSC hopefuls, leveraging their past execution information. This framework points to optimizing the learning involvement and upgrading the general victory rate of hopefuls by giving them with personalized ponder materials and assets. The proposed framework utilizes Machine Learning calculations, particularly the Network Factorization strategy, to extricate important bits of knowledge from the verifiable execution information of UPSC competitors. By analyzing the perplexing connections between the aspirants' execution measurements and the differing instructive substance accessible, the framework can precisely foresee the inclinations and learning designs of personal clients. Through the successful utilization of this prescient examination, the framework encourages the creation of personalized learning ways, empowering hopefuls to center on regions of shortcomings and strengthen ranges of quality. Besides, the AECRS joins a versatile learning system that ceaselessly upgrades and refines its proposals based on real-time input and client intuition. This versatile instrument guarantees that the framework remains responsive to the advancing needs and learning advance of each hopeful. By powerfully altering the prescribed substance based on the aspirants' continuous execution and inclinations, the framework maximizes engagement and maintenance, subsequently cultivating a more productive and viable learning preparation. The exploratory comes about to illustrate the viability of the proposed framework in improving the learning results of UPSC hopefuls. By giving custom-made instructive substance proposals, the framework enables hopefuls to optimize their consider techniques, in this manner expanding their chances of victory within the UPSC examinations. In general, this inquiry contributes to the headway of personalized learning advances within the field of competitive exam arrangement, laying the establishment for a more versatile and productive instructive environment.

In a one-size-fits-all conventional teaching context, student disparities in the background of pre-knowledge, skills, or comprehension lead to the alienation of the struggling students while boring those who are experienced. At issue is the appropriateness of the one-size-fits-all pedagogical model. An alternative is adaptive learning. Dated back to 1912, Edward L. Thorndike proposed the idea of a mechanical miracle that intends to conduct personal instructions through special print. This idea has inspired century-old efforts to automate education by creating such a kind of teaching machine that adaptively features automation, timely feedback, and self-pacing to student mastery in classrooms. With the technology evolving, a great range of state-of-the-art works that provide digital "teaching machines" have emerged, allowing adaptive learning. However, most of them focus on math or elementary reading and writing skills, and few are on programming-based courses. The few available are costly, heavy-weighted in adaptability, or limited in mapping competencies to an entire formal assessment, with questions that are not naturally differentiated in an assessment and thus, lacking flexibility. In response to these issues, we initialized an exploratory project and developed an agile adaptive learning web tool that maximizes the adaptive and connected learning experience of students on Canvas. This demo provides a brief tutorial on the steps to set up learning concepts, learning objectives, and objective-to-question mappings as well as on-the-fly quiz generation that matches required learning outcomes along with visualization of assessment outcome coverage and student mastery map. A teaching case will also be demonstrated. In the demo, we will use CS2-DS (Data Structures and Algorithms) as an example, but the ideas apply generally.

Recent advanced AI technologies, especially large language models (LLMs) like GPTs, have significantly advanced the field of data mining and led to the development of various LLM-based applications. AI for education (AI4EDU) is a vibrant multi-disciplinary field of data mining, machine learning, and education, with increasing importance and extraordinary potential. In this field, LLM and adaptive learning-based models can be utilized as interfaces in human-in-the-loop education systems, where the model serves as a mediator among the teacher, students, and machine capabilities, including its own. This perspective has several benefits, including the ability to personalize interactions, allow unprecedented flexibility and adaptivity for human-AI collaboration and improve the user experience. However, several challenges still exist, including the need for more robust and efficient algorithms, designing effective user interfaces, and ensuring ethical considerations are addressed. This workshop aims to bring together researchers and practitioners from academia and industry to explore cutting-edge AI technologies for personalized education, especially the potential of LLMs and adaptive learning technologies.

Large language models (LLMs) are revolutionizing the field of education by enabling personalized learning experiences tailored to individual student needs. In this paper, we introduce a framework for Adaptive Learning Systems that leverages LLM-powered analytics for personalized curriculum design. This innovative approach uses advanced machine learning to analyze real-time data, allowing the system to adapt learning pathways and recommend resources that align with each learner's progress. By continuously assessing students, our framework enhances instructional strategies, ensuring that the materials presented are relevant and engaging. Experimental results indicate a marked improvement in both learner engagement and knowledge retention when using a customized curriculum. Evaluations conducted across varied educational environments demonstrate the framework's flexibility and positive influence on learning outcomes, potentially reshaping conventional educational practices into a more adaptive and student-centered model.

The increasing complexity of educational challenges in technical disciplines highlights the need for personalized learning systems to address diverse student needs. Traditional methods, often relying on static activities or predefined rules, limit their ability to adapt to individual progress, hindering the development of critical skills such as problem-solving. Based on rules or machine learning, existing adaptive systems offer varying levels of personalization and efficiency but face significant scalability and computational demand barriers. This study proposes an adaptive learning system powered by deep learning algorithms designed to optimize problem-solving skills in technical college students. The system dynamically adjusts the difficulty of activities based on real-time performance data, ensuring a personalized and practical learning experience. A controlled experimental study was conducted with 200 students over eight weeks, divided into pretest, intervention, and posttest phases. The experimental group, which used the adaptive system, showed a 14% improvement in precision (from 71.8% to 85%) compared to 5% for the control group. In addition, the experimental group reduced its average time per activity by 15%, achieving 105 seconds compared to 124 seconds for the control group. These results demonstrate the system’s ability to improve precision, efficiency, and motivation in problem-solving tasks. By balancing computational efficiency with high personalization, this proposal offers a scalable and innovative solution that responds to current limitations in adaptive learning technologies.

No abstract available

No abstract available

Emerging technologies are transforming education, necessitating research on their optimal integration. This article introduces an Intelligent Virtual Reality (IVR) approach that incorporates Generative Artificial Intelligence (GAI) through two GAI-powered pedagogical characters, aiming to address educational needs. This qualitative descriptive research had two phases: Needs Analysis and Pedagogical Design. The needs and ideas, collected from 66 educators, were classified into three categories: AI acting as a character, emerging technologies to assist student learning, and emerging technologies to help teachers. The Pedagogical Design involved integrating GAI and VR in a sustainability education setting, and developing two virtual characters, Tero and Madida, as an information source and a learning companion, respectively. Evaluation through iterative testing with domain experts and interviews suggested that these characters met 9 out of 12 identified needs, highlighting their potential to enhance higher education learning experiences. Future research could explore further refinements to address the remaining needs.

Abstract Generative artificial intelligence (GenAI) represents a major leap forward in AI technology, offering the potential to reshape education in various aspects. This paper explores the transformative potential of GenAI in geography education, focusing on its impacts across curriculum, pedagogy, assessment, and fieldwork, through the lens of the Substitution, Augmentation, Modification, and Redefinition (SAMR) model. In curriculum development, GenAI enables automatic generation and personalization of geographic content. Pedagogical approaches are evolving from text-based instruction to data-driven learning experiences where students analyze geographic phenomena using GenAI tools. Assessment methods are shifting to adaptive evaluation systems with continuous feedback, while fieldwork benefits from real-time data processing and opportunities for global collaboration. Nevertheless, these advancements are accompanied by substantial risks, including challenges such as overreliance on AI, content inaccuracies, biases, and data privacy concerns.

The rapid advancement of generative AI technologies has opened new avenues for enhancing pedagogy in Engineering education. This paper explores the integration of generative AI into the teaching and learning processes, focusing on its potential to transform traditional pedagogical methods. By leveraging AI-driven tools, educators can create personalized learning experiences, automate routine tasks, and provide students with immediate feedback. The study investigates the impact of generative AI on student engagement, comprehension, and skill acquisition in core Engineering subjects. Through a series of case studies and empirical analysis, the research demonstrates how AI can be utilized to generate customized learning materials, facilitate interactive coding environments, and support collaborative learning at RK University. The findings suggest that incorporating generative AI into the curriculum not only enhances educational outcomes but also better prepares students for the evolving demands of the tech industry. This paper concludes with recommendations for educators on effectively implementing generative AI in Engineering programs and discusses the ethical considerations and challenges associated with its use in academic settings.

The interdisciplinary research domain of Artificial Intelligence in Education (AIED) has a long history of developing Intelligent Tutoring Systems (ITSs) by integrating insights from technological advancements, educational theories, and cognitive psychology. The remarkable success of generative AI (GenAI) models has accelerated the development of large language model (LLM)-powered ITSs, which have potential to imitate human-like, pedagogically rich, and cognitively demanding tutoring. However, the progress and impact of these systems remain largely untraceable due to the absence of reliable, universally accepted, and pedagogy-driven evaluation frameworks and benchmarks. Most existing educational dialogue-based ITS evaluations rely on subjective protocols and non-standardized benchmarks, leading to inconsistencies and limited generalizability. In this work, we take a step back from mainstream ITS development and provide comprehensive state-of-the-art evaluation practices, highlighting associated challenges through real-world case studies from careful and caring AIED research. Finally, building on insights from previous interdisciplinary AIED research, we propose three practical, feasible, and theoretically grounded research directions, rooted in learning science principles and aimed at establishing fair, unified, and scalable evaluation methodologies for ITSs.

This exploratory study looks at the pedagogical application of generative AI in architectural design studios. While most efforts focus on design automation using image synthesis function, this often reduces designers to passively selecting AI outputs. Instead, this study investigates alternative interactions between designers and AI and analyzes its pedagogical role in studio settings. Several case studies using building typologies demonstrate how generative AI can potentially expand students’ understanding of design concepts. A deep learning tool was used to analyze existing building types and generate new variations, allowing students to engage actively in design exploration. The study evaluates these outcomes, highlighting the dynamics of student-AI interactions and discussing the educational potential and limitations of using AI in design pedagogy.

ABSTRACT This article explores innovative strategies for educators to harness the potential of generative AI in writing instruction. Specifically, it focuses on pedagogical approaches teachers can use to help students navigate the complex landscape of AI while supporting their creative and critical thinking and instilling a firm ethical foundation in using AI. We anchor the strategies in the well-established educational theory of cognitive apprenticeship.

Artificial intelligence (AI) in general, and generative AI in particular, have already shown great potential as a transformative force across many disciplines, from medicine to education. In architecture and engineering education, generative AI may be viewed as an enabling method ology to provide students with enhanced learning, creativity, and problem-solving opportunities. This paper reviews how combining generative AI with traditional pedagogical methods could enhance innovation and educational outcomes. This study will analyze current applications, challenges and strategies for integrating generative AI into architecture and engineering curricula to show the benefits of this synergy while discussing the barriers to adoption. The findings emphasize adaptive teaching methods, ethical considerations, and upskilling to fully leverage the power of generative AI in reshaping education.

The present paper suggests an all-encompassing model of applying the Generative Artificial Intelligence in art education in the form of Hybrid Pedagogical Model of AI-Art Learning (HPM-AI). The model reformulates creativity as a collaborative activity between the human intent and algorithmic production, focusing on reflective learning, ethical consciousness and open participation. The analysis of the study brings together the cognitive theory, computational modeling, as well as case-based validation, which is why the study proves that generative AI can improve the diversity of creativity but without cultural and moral responsibility. Three contexts of implementation that included an AI-based design studio, a course on digital heritage restoration, and community-based art workshops were evaluated to determine the flexibility of the framework and its effect on pedagogy. Quantitative visualizations and qualitative thoughts show that HPM-AI contributes to different educational goals in settings: the conceptual innovation of studio learning, the cultural integrity of the practice of restoration, and the inclusive creative empowerment of the learning environment. The paper also discusses implications on authorship, data ethics, and psychology of the learner, that AI should not be viewed as a substitute to artistic skill, and instead it is an intelligent companion, which reflects and magnifies human imagination. The paper concludes, that the ethical design of generative AI in creative education, interdisciplinary faculty development and institutional transparency are the necessary conditions of sustainable integration. HPM-AI framework, therefore, promotes an image of symbiotic creativity, placing art pedagogy on a crossroad of human cognition in relation to cultural continuity in connection with the computational intelligence.

This paper explores the paradigm reconstruction of interpreting pedagogy driven by generative AI technology. With the breakthroughs of AI technologies such as ChatGPT in natural language processing, traditional interpreting education faces dual challenges of technological substitution and pedagogical transformation. Based on Kuhn’s paradigm theory, the study analyzes the limitations of three traditional interpreting teaching paradigms, language-centric, knowledge-based, and skill-acquisition-oriented, and proposes a novel “teacher-AI-learner” triadic collaborative paradigm. Through reconstructing teaching subjects, environments, and curriculum systems, the integration of real-time translation tools and intelligent terminology databases facilitates the transition from static skill training to dynamic human-machine collaboration. The research simultaneously highlights challenges in technological ethics and curriculum design transformation pressures, emphasizing the necessity to balance technological empowerment with humanistic education.

Generative Artificial Intelligence (AI) technologies such as ChatGPT are rapidly transforming global educational practices in the era of digital innovation. However, their pedagogical application within rural and religious educational settings remains underexplored. Grounded in the Technological Pedagogical Content Knowledge (TPACK) and Diffusion of Innovation (DOI) frameworks, this study investigates how generative AI influences pedagogical strategies among teachers in Islamic secondary schools in Southern Pakistan, particularly in the Kot Addu district. The research aims to: (1) determine the level of ChatGPT and AI adoption among teachers; (2) examine how generative AI impacts pedagogy, lesson planning, and student interaction; and (3) analyze gender-based differences in perceptions and utilization of these tools. A quantitative research design was employed using a structured questionnaire administered to 440 Islamic secondary school teachers. Data were analyzed using SPSS 2021 through descriptive statistics, t-tests, and correlation analysis. The findings reveal moderate adoption of AI technologies, which positively contribute to instructional innovation and teacher–student engagement. Notably, male teachers demonstrated slightly higher confidence and more active use of generative AI tools compared to female teachers. The study acknowledges limitations such as reliance on self-reported data, infrastructural constraints related to internet and device access, and cultural reservations regarding AI integration in religious education. The findings highlight the importance of targeted teacher training, culturally sensitive digital policies, and institutional support to ensure equitable integration of AI. Future research should incorporate qualitative approaches to capture deeper insights into teachers’ experiences, examine long-term impacts on learning outcomes, and explore policy-level enablers and barriers to AI integration in Islamic educational contexts.

In an increasingly globalized and linguistically diverse world, English Language Teaching (ELT) must evolve to address the needs of all learners, including those from marginalized, multilingual, and neurodiverse backgrounds. The integration of Generative Artificial Intelligence (AI) into language pedagogy marks a significant paradigm shift toward inclusivity, personalization, and accessibility. This paper explores how generative AI technologies such as ChatGPT, Grammarly, ELSA Speak, and other adaptive platforms can support inclusive ELT practices by aligning with frameworks like Universal Design for Learning (UDL), translanguaging pedagogy, and culturally responsive teaching. This paper highlights how AI tools are democratizing English education, especially in resource-constrained environments. However, the paper also critically examines the ethical and pedagogical challenges associated with AI, including algorithmic bias, overreliance, privacy concerns, and the digital divide. By addressing both the affordances and limitations of generative AI, this study underscores the importance of thoughtful integration strategies grounded in equity and learner empowerment. The research ultimately advocates for a hybrid model of AI-enhanced pedagogy that preserves the irreplaceable role of human teachers while embracing the transformative potential of AI. The findings aim to contribute to ongoing discussions around sustainable, ethical, and inclusive practices in 21st-century English language education.

The proliferation of Generative AI necessitates a re-evaluation of educational strategies, particularly in vocational fields. Traditional vocational education faces challenges like limited resource access, high software costs, and a lack of personalized feedback. This paper explores how integrating generative AI, guided by a human-centered philosophy, can address these issues. Through a qualitative analysis of four pedagogical interventions at a vocational school (e-commerce, art, math, and computer science), we find that AI, as a pedagogical co-pilot, boosts instructional efficiency, nurtures creativity, and enables individualized learning. The case studies show AI's ability to lower costs, remove practice barriers, and provide data-driven insights. We synthesize these findings into a conceptual framework for human-centered AI integration, emphasizing AI's role in empowering educators and learners. This research offers a transferable model and discusses ethical considerations for creating effective and equitable learning environments.

No abstract available

No abstract available

Despite increased recognition of the importance and need for pedagogical training for public health and health promotion instructors in best-practices and inclusivity, formal training is often overlooked. This disregard for pedagogical training necessitates exploration of alternative and innovative approaches to enhance teaching and learning such as generative AI. This paper describes applied uses of generative AI, specifically ChatGPT, to enhance pedagogy in public health and health promotion education in the areas of curriculum design, instructional strategies, assessment and feedback, and diversity, equity, and inclusion. Generative AI as a supplemental tool shows immense promise for improving teaching and learning, however, inherent limitations and ethical considerations require caution and continued scrutiny.

This article offers a semester-long approach to using generative AI in the public-speaking course. Using critical communication pedagogy, the authors provide practices to navigate the turbulence that has followed the emergence of publicly available generative AI tools. These tools have received negative attention because of their potential to facilitate academic dishonesty and undermine the learning process. Despite this concern, we’ve chosen to embrace these tools and the transformative power they have for public-speaking curricula. Course Public Speaking. Objectives Create an environment of co-learning with students; formulate expectations for student use of AI; develop a reflective process for students to critically evaluate AI material; and design assignments to encourage critical thinking skills.

This paper explores a novel workflow that integrates Generative AI tools, ChatGPT and DALL·E, into educational use, aiming to improve the traditional teaching methods in university education. Our workflow is focused on creating short introductory videos for university courses, using primary course descriptions available in the university’s study guide with the idea of introducing courses visually. This approach was deliberately selected for experimentation, and we believe that it could be further enhanced to generate course videos on specific course topics. This will minimize the efforts of teachers who are required to produce detailed course videos as a part of their teaching. As the first part of our workflow, we present a tool that utilizes ChatGPT-4 and DALL·E 2 to autonomously generate a script and background graphics for videos, using primary course descriptions extracted through a given course web URL. As the second part of the workflow, we combine those generated artefacts into videos using Narakeet, a Text-to-Speech software service that is available online. To analyze the feasibility of this workflow, we then conducted a field survey where university teachers participated in reviewing introductory course videos of their courses generated through our workflow. We employed only engineering courses that are English-taught in this field survey. The results demonstrate the potential of AI-generated content to increase the efficiency of teachers when creating video materials. However, challenges such as the uncanny valley effect in text-to-speech narration and the propensity for AI-generated misinformation highlight the need for careful review by humans on such content before setting it for wider use. This paper argues for the strategic integration of AI in university education, focusing on the benefits, while acknowledging the limitations owned by generative AI tools.

This study employs generative AI to revamp a core engineering course, thermodynamics, to boost student engagement and comprehension. In collaboration with faculty, students, and AI specialists, the effort explores effective AI tools and strategies for question generation and the creation of digital aids, promising a significant impact on student participation and overall learning experience. This approach not only enhances learning experiences but also fosters a culture of innovation, suggesting significant potential for applying these methods across various courses. The initiative aims to refine pedagogical practices through strategic AI tool integration, highlighting the evolving role of technology in education. The study integrates AI models for educational content, employing advanced AI like GPT-3.5, 4, Gemini, DALL-E, Eduaide, and Llama 2. It fine-tunes AI settings for optimal performance and rigorously assesses the quality of generated content and images, revealing AI's potential in creating relevant educational materials. However, challenges in accurate visual representation persist.

No abstract available

This paper explores the synergy between human cognition and Large Language Models (LLMs), highlighting how generative AI can drive personalized learning at scale. We discuss parallels between LLMs and human cognition, emphasizing both the promise and new perspectives on integrating AI systems into education. After examining challenges in aligning technology with pedagogy, we review AutoTutor-one of the earliest Intelligent Tutoring Systems (ITS)-and detail its successes, limitations, and unfulfilled aspirations. We then introduce the Socratic Playground, a next-generation ITS that uses advanced transformer-based models to overcome AutoTutor's constraints and provide personalized, adaptive tutoring. To illustrate its evolving capabilities, we present a JSON-based tutoring prompt that systematically guides learner reflection while tracking misconceptions. Throughout, we underscore the importance of placing pedagogy at the forefront, ensuring that technology's power is harnessed to enhance teaching and learning rather than overshadow it.

A major challenge facing the world is the provision of equitable and universal access to quality education. Recent advances in generative AI (gen AI) have created excitement about the potential of new technologies to offer a personal tutor for every learner and a teaching assistant for every teacher. The full extent of this dream, however, has not yet materialised. We argue that this is primarily due to the difficulties with verbalising pedagogical intuitions into gen AI prompts and the lack of good evaluation practices, reinforced by the challenges in defining excellent pedagogy. Here we present our work collaborating with learners and educators to translate high level principles from learning science into a pragmatic set of seven diverse educational benchmarks, spanning quantitative, qualitative, automatic and human evaluations; and to develop a new set of fine-tuning datasets to improve the pedagogical capabilities of Gemini, introducing LearnLM-Tutor. Our evaluations show that LearnLM-Tutor is consistently preferred over a prompt tuned Gemini by educators and learners on a number of pedagogical dimensions. We hope that this work can serve as a first step towards developing a comprehensive educational evaluation framework, and that this can enable rapid progress within the AI and EdTech communities towards maximising the positive impact of gen AI in education.

Artificial intelligence (AI) models struggle to reach performance levels due to the complex nature of Arabic grammar and diverse regional dialects. This study investigated how generative AI (GenAI) functions as a teaching assistant in Arabic language classrooms. Using qualitative methods, semi-structured interviews were conducted with 15 instructors; the data was then analyzed using thematic analysis. Results revealed that instructors used GenAI to create material, assess students' work, and create personalized learning plans. Instructors struggled, however, with AI accuracy in dialect processing, cultural authenticity, and ensuring accurate assessment methods. The analysis raised significant gaps in teacher training, assessment strategies, and institutional guidelines. Instructors found it challenging to evaluate AI-generated Arabic content across different dialects and maintain academic integrity in student assignments. This study recommends developing instructor training, specifically on using GenAI tools for Arabic dialect variations and creating culturally appropriate Arabic language learning materials.

For today's Net generation (especially the Alpha generation born between 2010 and 2024) and all subsequent generations, it will be imperative to redefine educational strategies in the context of rapid technological and societal change. It is necessary to consider how to proceed, as the educational challenge must be placed in a broader philosophical and cultural context, with an emphasis on the ever-rapidly evolving technology as well as the nature of knowledge and human experience. Based on the paradigm of the shift from Web 2.0 to Web 4.0 and the consequent shift from Education 4.0 to Education 5.0, potential guidelines for the development of modern education 4.0 based on digital pedagogy that combines personalized learning, real-time feedback and collaborative, interdisciplinary environments in Education 5.0 are indicated. This reflection will place particular emphasis on the role of teachers as mentors and not merely transmitters of information, as well as on the ethical, social and emotional dimensions of digital learning, and emphasize the importance of adapting educational practices to real-life contexts and future humanistic challenges of education (Flogie et al., 2025).

This paper reconceptualizes the role of the teacher in the university foreign language classroom in an age of generative AI chatbots and automatic translation tools. We call for a reconceptualization of this role based on two factors: the unique social interactivity of the university language classroom and the need for effective instruction on how to use Intelligent Computer-Assisted Language Learning (ICALL) tools outside of the classroom. We argue that the teacher must master and integrate these two different modes of teaching and learning. Interpersonal exchanges in class respond to the need for real-time human interaction and relatedness in language learning and so cannot, and should not, be wholly replaced by chatbots. Rather, these sorts of exchanges must form a cornerstone of on-campus foreign language pedagogy. In contrast, teachers must also be able to leverage the benefits of learner-facing AI tools, especially for use outside of the classroom, given the learning gains associated with them. We provide detailed examples of how this dual approach can be realized and propose a five-step approach for incorporating AI into university language pedagogy.

This study examines the transformative role of generative AI (e.g., ChatGPT, Grammarly) in academic writing education, focusing on its dual capacity to enhance technical proficiency while challenging originality and critical thinking. Drawing on a mixed-methods analysis of 300 undergraduates and 45 educators across eight universities, this paper reveals that structured AI integration improves grammar and citation accuracy by 32% but correlates with a 19% decline in argument originality. By synthesizing Vygotskian scaffolding theory with posthumanist pedagogy, we propose a co-creative framework emphasizing transparency, phased tool usage, and adaptive assessment to preserve human agency in the AI era.

No abstract available

educational Aim/Background: The possible lack of adaptable and effective student support systems in conventional educational techniques may hinder the continuous development of successful academic learning process, and lead to inconsistent learning outcomes. Generative AI chatbots have the potential to change pedagogy and learning environment by influencing students' academic practices and personalized experiences. This study aims to present a novel generic step-by-step framework based on competency-based learning (CBL) approach, to improve student academic practices using generative AI-powered chatbots, Methods: The proposed roadmap framework integrates the competency structure learning methodology with chatbot tools; and provides appropriate guided prompt examples for each part, related to the “Industrial Automation” engineering course as a guided subject. The targeted goal is to enhance students', knowledge, skills, and attitudes (KSAs); hence boosting overall learning outcomes. The effectiveness has been evaluated through three different students’ groups (A, B and C), across three different courses, using mixed quantitative surveys, questionnaires and qualitative creative tasks. Results: Quantitative data from surveys and questionnaires, along with qualitative responses to the creative tasks, indicated student positive perceptions and self-response. Results analysis of the three groups showed the enhancement of student engagement and enabling to explore topics on their own pace and receiving personalized targeted feedback. Conclusion: This paper has suggested a promising generic framework, integrating the competency-based learning structure approach with generative AI chatbots, to enhance student academic practices. Practical "Do" and "Don't" advice to help learners to minimize possible associated problems, such as over-reliance and plagiarism is finally listed.

Generative AI presents a profound challenge to the existing structures and purposes of education. It forces us to reconsider not only how we teach and learn but also, more fundamentally, what education is for. This conceptual paper argues that, in order to integrate AI into education in a way that can meet the major challenges facing humanity, ranging from ecological crisis to the future of democratic societies, we must reframe education. Drawing on the nature and potential of generative AI in conjunction with educational theory, we propose a double dialogic pedagogy that recognises education as both teaching thinking through dialogue and inducting students into participation in the long‐term powerful dialogues of culture. We relate this double dialogic pedagogy with AI to education for collective intelligence. This pedagogy positions AI not as a replacement for human thinking, but as a partner in expanding the space of dialogue. By articulating this theoretical foundation, we offer a basis for the future design of educational practices and technologies that can support human flourishing in an age of accelerating technological change. What is already known about this topic The rise of generative AI, particularly large language models, poses both opportunities and challenges for education. While existing research highlights AI's potential to support personalised learning through intelligent tutoring systems, concerns remain about its possible negative impact on students' critical and creative thinking. Previous studies have also noted the need for educational theory to keep pace with technological developments, yet little work has been done to articulate a pedagogical framework that addresses the broader societal and environmental challenges of the AI era. What this paper adds This paper proposes a double dialogic pedagogy as a theoretical foundation for integrating AI into educational design. It situates current developments in AI within a historical and philosophical context, drawing on dialogic theory and the concept of the pharmakon to argue that AI is neither inherently beneficial nor harmful, but depends on pedagogical framing. The paper advances the field by explicitly linking AI‐supported learning to education for collective intelligence, offering concrete illustrations of how AI can support both dialogic learning and induction into long‐term cultural dialogues. Implications for practice and/or policy This paper suggests that educational practice should shift from focusing on individual knowledge acquisition towards cultivating collective intelligence through dialogic pedagogy. Policymakers and educators are encouraged to design AI‐supported learning environments that promote collaborative problem‐solving and critical engagement with real‐world challenges, such as those posed by the Anthropocene. Assessment policies should be re‐evaluated to reward collaborative inquiry and ethical reasoning, rather than individual content reproduction. This reconceptualisation positions education as a key enabler of planetary‐level self‐regulation and human flourishing in the face of technological and environmental disruptions.

As Artificial Intelligence (AI) becomes more deeply embedded in educational settings, it is crucial to explore how it can enhance—rather than replace—the role of educators. This workshop focuses on advancing AI-driven tools that support personalized learning and designing AI systems capable of understanding students’ emotional and cognitive needs. The workshop will explore two main themes: (1) empowering teachers with AI technologies for delivering customized feedback and individualized instruction, and (2) examining ethical and practical considerations for developing empathetic AI that complements the human aspects of teaching. Participants will discuss opportunities and challenges in building effective AI-augmented learning environments, considering ethical concerns such as privacy, equity, and the maintenance of human agency. The workshop aims to unite educators, researchers, and technologists in developing innovative solutions and fostering interdisciplinary dialogue. Key outcomes include establishing best practices for AI integration in education and disseminating research findings that shape the future of human-AI collaboration in teaching.

No abstract available

No abstract available

In service of the goal of examining how cognitive science can facilitate human–computer interactions in complex systems, we explore how cognitive psychology research might help educators better utilize artificial intelligence and AI supported tools as facilitatory to learning, rather than see these emerging technologies as a threat. We also aim to provide historical perspective, both on how automation and technology has generated unnecessary apprehension over time, and how generative AI technologies such as ChatGPT are a product of the discipline of cognitive science. We introduce a model for how higher education instruction can adapt to the age of AI by fully capitalizing on the role that metacognition knowledge and skills play in determining learning effectiveness. Finally, we urge educators to consider how AI can be seen as a critical collaborator to be utilized in our efforts to educate around the critical workforce skills of effective communication and collaboration.

Artificial intelligence is poised to transform teaching and coaching practices,yet its optimal role alongside human expertise remains unclear.This study investigates human and AI collaboration in fitness education through a one year qualitative case study with a Pilates instructor.The researcher participated in the instructor classes and conducted biweekly semi structured interviews to explore how generative AI could be integrated into class planning and instruction.

This study investigates the integration of AI-assisted dialogue and image-literate instruction in primary art education, grounded in neuroscience models of multisensory information processing. Drawing inspiration from the visual-auditory convergence in the lateral and medial geniculate nuclei, the instructional design leverages Doubao—a conversational AI with voice output—to iteratively engage students in decoding traditional Chinese artworks. Conducted over four weeks with 253 sixth-grade students across six classes, the curriculum focused on selected works from the painting Lady Guoguo's Spring Outing, a classic from the Tang Dynasty and one of the Top Ten Chinese Paintings. Students were guided to observe visual elements, inquire into unfamiliar cultural symbols, and refine their interpretations through voice-augmented AI dialogue. The system's responsive feedback loop encouraged successive questioning, style transfer-based imaginative tasks, and repeated information verification. This process enhanced students’ curiosity and strengthened their ability to evaluate information reliability by layering and comparing contextual cues. Results indicate that combining image and sound through human-AI collaboration fosters deeper engagement, cultural literacy, and metacognitive awareness in visual art learning.

No abstract available

Text prompt is the most common way for human-generative AI (GenAI) communication. Though convenient, it is challenging to convey fine-grained and referential intent. One promising solution is to combine text prompts with precise GUI interactions, like brushing and clicking. However, there lacks a formal model to capture synergistic designs between prompts and interactions, hindering their comparison and innovation. To fill this gap, via an iterative and deductive process, we develop the Interaction-Augmented Instruction (IAI) model, a compact entity-relation graph formalizing how the combination of interactions and text prompts enhances human-GenAI communication. With the model, we distill twelve recurring and composable atomic interaction paradigms from prior tools, verifying our model's capability to facilitate systematic design characterization and comparison. Four usage scenarios further demonstrate the model's utility in applying, refining, and innovating these paradigms. These results illustrate the IAI model's descriptive, discriminative, and generative power for shaping future GenAI systems.

With the rapid advancement of artificial intelligence, particularly in natural language processing, the field of education is undergoing profound transformation. As a critical component of language education, Chinese writing instruction—characterized by its complex cognitive structure and linguistic features—offers a multilayered space for technological integration. This paper systematically reviews the theoretical foundations, technological approaches, and adaptive conditions for applying AI in Chinese writing instruction. It explores the construction of instructional support systems centered on language models, semantic analysis, and human–machine collaboration, and evaluates their feasibility and practical value from three perspectives: learners’ cognitive characteristics, types of instructional content, and the division of roles between teachers and technology. The findings suggest that AI-assisted systems demonstrate significant potential in language error correction, structural optimization, and process-oriented feedback, thereby enhancing instructional efficiency and learners’ writing proficiency. Future efforts should focus on improving contextual awareness, cross-task generalization, and mechanisms for human–AI collaboration to further advance the intelligent transformation of Chinese writing instruction.

This essay uses a loose auto-ethnographic approach to suggest ways to use artificial intelligence (AI) in teaching college English writing in the Chinese context. We suggest integration of AI technology into the entire writing process –brainstorming, drafting, and revision– with particular attention to challenges faced in Chinese education. We also explore how AI can assist teachers with innovative instructional planning, student performance analysis, and evaluation, reducing the burden of repetitive work and allowing teachers to focus on cultivating students’ higher-order thinking abilities, thus realizing a new human-machine collaboration teaching paradigm. The next step will be to formally execute these steps in a full research model with control groups. We would particularly welcome researchers from other universities and other cultural contexts to join us in this exploration.

Contribution: This Full paper in the Research Category track describes a practical, scalable platform that seamlessly integrates Generative AI (GenAI) with online educational forums, offering a novel approach to augment the instructional capabilities of staff. The platform empowers instructional staff to efficiently manage, refine, and approve responses by facilitating interaction between student posts and a Large Language Model (LLM). Background: This study is anchored in Vygotsky's socio- cultural theory, with a particular focus on the concept of the More Knowledgeable Other (MKO). It examines how GenAI can augment the instructional capabilities of course staff in educational environments, acting as an auxiliary MKO to facilitate an enriched educational dialogue between students and instructors. This theoretical backdrop is important for understanding the integration of AI within educational contexts, suggesting a balanced collaboration between human expertise and artificial intelligence to enhance the learning and teaching experience. Research Question: How effective is GenAI in reducing the workload of instructional staff when used to pre-answer student questions posted on educational discussion forums? Methodology: Employing a mixed-methods approach, our study concentrated on select first and second-year computer programming courses with significant enrollments. The investigation involved the use of an AI -assisted platform by designated (human) Teaching Assistants (AI- TAs) to pre-answer student queries on educational forums. Our analysis includes a qualitative examination of feedback and interactions, focusing on the AI-TAs' experiences and perceptions. While we primarily analyzed efficiency indicators such as the frequency of modifications required to AI generated responses, we also explored broader qualitative aspects to understand the impact and reception of AI -generated responses within the educational context. This approach allowed us to gather insights into both the quantitative engagement with AI -assisted posts and the qualitative sentiments expressed by the instructional staff, laying the groundwork for further in-depth analysis. Findings: The findings indicate no significant difference in student reception to responses generated by AI - TAs compared to those provided by human instructors. This suggests that GenAl can effectively meet educational needs when adequately managed. Moreover, AI - TAs experienced a reduction in the cognitive load required for responding to queries, pointing to GenAI's potential to enhance instructional efficiency without compromising the quality of education.

Artificial Intelligence (AI) presents transformative potential for human-centered design education by enabling personalized learning, enhancing collaboration, and expanding access to tailored resources.  This study aimed at exploring AI technologies support for instruction in design education, especially their application in personalization, online collaborative learning, and equitable access, along with their ethical considerations. A mixed-methods approach was utilized in this study that included surveys combined with semi-structured interviews and classroom observations. The purposive sample of 60 participants was made up of 25 instructors, 30 students, and 5 design professionals at Stephen F. Austin State University. Quantitative data were analyzed using descriptive statistics and Pearson correlation, toward identifying correlations between learner performance and the usage of AI, while qualitative data was thematically analyzed using NVivo. Results showed how AI significantly facilitates individualized learning through real-time feedback and personalized content (Demszky et al., 2023; Woolf, 2010), facilitates collaboration through intelligent platforms (Suthers, 2006), and opens access to relevant resources (Ferguson, 2012; Chiu et al., 2023). Data privacy issues, bias (Luckin et al., 2016; O’Neil, 2016), and the ability to augment educational inequity (Warschauer, 2004) issues were raised, however. The study concludes on the premise that AI can support effective learning and inclusion if complemented by ethical surveillance and access parity interventions

The rapid advancement of Artificial Intelligence in education (AIEd) presents unprecedented opportunities to improve learner engagement through personalized instruction and human-robot social interaction (HRSI). However, AI implementation in preservice teacher education in Indonesia remains limited and underexplored. This study investigates the impact of AI-powered personalization on learning outcomes and social interactions among Indonesian preservice teachers. Employing a mixed-method design, the study involved 20 participants using the virtual AI tutor "Cicibot" to support personalized and collaborative learning. Quantitative data were collected via structured questionnaires and analyzed using Pearson correlation, while qualitative insights were obtained from semi-structured interviews. Findings reveal a significant and positive correlation between AI-driven personalization, learner engagement, and social interaction, highlighting the effectiveness of AI tools in fostering meaningful collaboration. This study provides practical implications for AI integration in educational settings, offering insights for future policy and curriculum development in technologically emergent regions.

This research was conducted to investigate AI integration in the development and evaluation of English oral communication skills, within both academic and professional practices. This study used a mixed-methods design, combining data from a quasi-experiment with qualitative data collected through interviews with students and instructors. Results demonstrated that the AI-infused instruction and assessment significantly improved English oral communication in fluency, accuracy, and general proficiency. Thematic analysis of the interviews indicated that AI technologies can provide individualized feedback, practice opportunities, and objective assessment in the promotion of effective language learning. However, these processes raise concerns about AI bias, human-AI collaboration, and cultural and pedagogical contexts. The recommendations of this research provided some actionable points for educators and program administrators on strategically incorporating AI tools in ways that will positively affect English-speaking proficiency.

This paper examines the deep application of artificial intelligence (AI) in higher education, focusing on teaching reform across economics, management, and the humanities. To address persistent challenges—namely, the scarcity of replicable cases, the disconnect between theory and practice, slow curriculum renewal, and weak value guidance—we propose a systemic reform pathway. Grounded in human‑centered educational theory, the approach integrates competency-oriented curriculum redesign, innovation in project-based teaching, an intelligent closed-loop assessment system, multi‑dimensional empirical validation, and phased, system‑level governance. Together, these elements support the organic integration of AI and humanistic education. The core objective is to cultivate three key competencies: data literacy, model literacy, and humanistic‑legal literacy, while building a modular curriculum system and promoting a shift from teacher‑centered instruction to project‑based learning that is problem‑driven, data‑informed, and collaboration‑oriented. Ultimately, the framework aims to provide a teaching‑reform paradigm that combines theoretical depth with practical value and serves as a model for the intelligent transformation of humanities and social‑science education.

The vigorous development of artificial intelligence (AI) represented by large language models (LLMs) has rapidly promoted the updating and development of educational technology. Agentic workflows (AWs) built based on LLMs can realize complex tasks in the field of education, which allows the emergence of swarm intelligence (SI) through multi-agent collaboration[1]. This study introduces the Agent4EDU (agent for education) framework, which outlines 4 application models in education from the two dimensions of degree of agency and degree of interaction, including human-AI collaboration, AI assistant, instruction execution, and general type. The proposed Agent4EDU framework discusses the paradigm of educational applications of AI agents and promotes the development of the field of AI for education.

This study proposes an AI-enhanced design thinking framework to improve knowledge creation and transfer in digital media education. As generative AI transforms creative practices, traditional instruction lacks adaptability and personalization. By integrating AI tools like large language models into design thinking, the framework enables dynamic knowledge construction, iterative prototyping, and intelligent feedback. It supports adaptive knowledge management systems in education, enhancing curriculum flexibility and teaching efficiency. Empirical results show improved knowledge sharing, reuse of learning artifacts, and higher learner satisfaction. Aligning AI with human-centered design fosters cognitive engagement and collaboration. Findings highlight AI's role as a cognitive and collaborative enabler in pedagogy. This research contributes to knowledge management by demonstrating how AI-driven methods can structure knowledge capture, transfer, and application in higher education. The framework offers a model for AI-integrated, learner-centered curriculum innovation.

Integrating social computing with a combination of AI improvement offers transformative potential for both instructing and exciting applications. This paper investigates how social computing-characterized by collaborative and intelligent client behavior in advanced environments-can be improved through AI advances to make more personalized, engaging, and versatile encounters. Within the instruction set, AI-driven social computing frameworks empower clever coaching, real-time input, and peer collaboration, cultivating higher levels of understudy engagement and personalized learning pathways. In amusement, AI-powered social stages can create immersive, intelligently encounters custom-fitted to person inclinations, advertising unused shapes of substance creation and client support. Be that as it may, whereas these innovations hold excellent guarantees, they, too, show critical challenges, such as guaranteeing client security, tending to moral concerns, and planning frameworks that viably adjust human and machine intelligence. This paper analyzes the benefits, potential applications, and challenges of blending social computing with AI within instruction and excitement, giving a system for future inquiry about and advancement in this advancing space.

Open access to novel AI tools offers unprecedented opportunities for human–AI collaboration in writing instruction and assessment. While research on using generative AI tools like ChatGPT in these contexts is emerging, more is needed to understand their effectiveness as Automated Writing Evaluation (AWE) tools. This study explores the potential of ChatGPT (GPT‐3.5) to assist teachers and learners in the North Atlantic Treaty Organization (NATO) by evaluating English writing based on holistic scoring criteria. Using a mixed‐methods approach, the study compared ChatGPT's ratings with human ratings on 100 writing tests to assess inter‐rater reliability. It also analyzed the justifications provided by both human raters and ChatGPT to evaluate how well ChatGPT understood the rating criteria at different proficiency levels and whether its rationales could provide effective feedback for learners and support teacher feedback practices. Results showed strong agreement between ChatGPT's and human ratings, with ChatGPT demonstrating a similar understanding of the rating scales and offering justifications with elements of effective feedback. These findings indicate that ChatGPT holds promise as an AWE tool, providing meaningful feedback and valuable insights into holistic rating scales. This study encourages further exploration of AI in the L2 classroom and suggests leveraging AI to enhance writing pedagogy and classroom‐based assessment.

This study explores strategies for integrating Generative Artificial Intelligence (GAI) into high school English writing instruction to address challenges such as limited personalized feedback and resource constraints in traditional teaching. Focusing on the integration of tools like ChatGPT and Kimi, the research employs a quasi-experimental mixed-methods design to evaluate the efficacy of GAI-empowered strategies. The study systematically examines three core dimensions: the design of a GAI-supported teaching framework spanning pre-writing, drafting, and revision phases; the implementation of stratified feedback mechanisms for personalized writing guidance; and the optimization of teacher-AI collaboration to balance automated support with human expertise. Quantitative data from pre/post-test writing assessments and Likert-scale surveys, combined with qualitative insights from classroom observations and interviews, reveal significant improvements in students’ writing accuracy, coherence, and motivation, while also identifying challenges such as content bias and the need for teacher training. The findings contribute to both theoretical discourse and practical pedagogy, offering a replicable model for aligning GAI with curriculum standards. This research not only underscores GAI’s role in writing instruction but also provides critical recommendations for sustainable technology integration in classrooms.

As artificial intelligence generated content (AIGC) technology continues to develop, its potential in higher education has attracted significant attention. This study examines the application of AIGC in English language teaching within tourism and cultural vocational undergraduate programs. Aligned with the College English Teaching Guidelines and the principles of vocational education, it explores how AIGC can enhance curriculum design, support scenario-based instruction, and strengthen professional competency development. Through practical examples including AI-powered dialogue simulation and industry-aligned text generation, the study highlights the technology’s effectiveness in increasing instructional relevance and enabling personalized learning. Challenges related to human-AI collaboration, ethical data use, and pedagogical adaptation are also discussed, offering insights for English teaching reform in vocational undergraduate education.

With the rapid advancement of artificial intelligence (AI), Large Language Models (LLMs) have garnered significant attention due to their robust content generation capabilities and user-friendly, practical interactive experiences. In the context of efficient education, LLMs are at the forefront of driving the evolution of digital and intelligent educational practices, facilitating the reduction of teaching burdens for educators while enhancing the quality and efficiency of instruction. First of all, this study discusses the application prospect of LLMs in supporting electronic information teaching, and shows that LLMs can play a greater role in enabling the teaching implementation link which is more specific and needs human-mechine collaboration. Then we use the problem-solving research method to analyze three key challenges inherent in electronic information experimental teaching within higher education institutions: the complexity of lesson preparation for instructors, the difficulties students face in comprehension, and the challenge of tailoring instruction to individual learning needs. In response to these identified issues, the paper proposes targeted strategies leveraging the comprehensive capabilities of LLMs to assist electronic information experimental teaching, thereby promoting improvements in both teaching quality and efficiency. Through the thinking and discussion of specific problems, it provides a new application idea for new artificial intelligence new technology enabling experimental classroom and improving teaching quality and efficiency.

With the rapid advancement of artificial intelligence technology, its potential for application in the field of education has become increasingly prominent. Secondary vocational schools have long faced challenges in Photoshop skill instruction, including significant disparities in student foundational knowledge, extended skill acquisition periods, and difficulties in fostering creativity. Grounded in constructivist learning theory and personalized learning theory, this study explores the theoretical basis and feasibility of using AI tools to enhance Photoshop skill acquisition. It constructs a teaching model centered on "human-machine collaboration, data-driven decision making, and dynamic adaptation," and designs specific implementation pathways from three perspectives: technology integration, task design, and assessment optimization. By establishing a multi-dimensional learning efficiency evaluation system, the study systematically analyzes learning outcomes following the integration of AI tools. The findings demonstrate that the introduction of AI tools such as Adobe Sensei and Remove.bg significantly improves the learning efficiency of secondary vocational students in areas including operational proficiency, workflow optimization, and creative expression. This research provides a transferable framework for innovating digital skill instructional models in vocational education.

This innovative practice full paper proposes Creator-Thon, a novel teaching project designed to introduce Generative AI into traditional interior design workflows. The project adopts an interdisciplinary approach including application scenario investigation, workflow construction, collaborative innovation, and iterative optimization. The underlying research is grounded in a six-level goal hierarchy, comprising sub-goals of integrating AI capabilities and design knowledge, embedding AI skills into workflows, enhancing the learning engagement, empowering personal career development, fostering (human-AI and human-human) collaboration, and transforming the design paradigms. The teaching framework is progressively refined across multiple iterations, enhancing learning engagement by simplifying tool operations, refining design language, and re-engineering the design process. The findings underscore that integration of authentic application scenarios markedly boosts engagement and motivation among learners, modular instruction also enhances teaching quality and practical skills, and the adaptive optimization of the teaching framework informed by participants' feedback, is key to a successful Generative AI involved interior design pedagogy. This work provides insights that contribute to the application of Generative AI technology in interior design education and practice, facilitating transformations and progress in AI literacy in the field.

No abstract available

No abstract available

Abstract: Effectiveness of Artificial Intelligence-Based Learning Analytics Tool in Supporting Personalized Learning in Higher Education. Objective: The use of Artificial Intelligence (AI) in higher education has become one of the significant innovations to improve the quality of learning, especially through personalization. However, the effectiveness of AI-based learning analytics tools in supporting learning personalization in Indonesian higher education is still not widely explored. This research aims to evaluate the effectiveness of AI-based analytics tools in supporting learning personalization and identify challenges in its implementation. Method: This research uses descriptive qualitative method with in-depth interviews, observation, and document analysis. The respondents included 10 lecturers and 15 students from three universities in Indonesia who have implemented AI-based tools in the learning process. Data were analyzed using thematic analysis techniques to identify key patterns and trends. Finding: The results show that AI-based analytics tools are effective in improving learning personalization. Students using the tool recorded an average academic grade improvement of 15.03%, while lecturers found it helpful in understanding students' learning needs. The main challenges identified were low digital competency (50% of respondents) and concerns regarding data privacy (30% of respondents). Conclusion: AI tools have great potential to support personalized learning, but their success depends on digital literacy training and more transparent data management. This research has implications for the importance of investing in technology-based education infrastructure and policies to ensure sustainable adoption. Keywords: artificial intelligence, personalized learning, higher education, technology effectiveness.

Personalized learning requires technology in order to feed information about the learners’ preferences, achievements and needs. However, many available works have focused either on fully automated processes such as prediction or recommendation of learning plan. In fact, integration of human into the machine learning loop would improve the performance of the assistive technology compared to separated models. In this paper we propose an approach for personalized learning that integrates human and machine learning, and utilize learning analytics, chatbot and recommendation system. We present our proposed idea and work in progress.

Learning analytics, as a rapidly evolving field, offers an encouraging approach with the aim of understanding, optimizing and enhancing learning process. Learners have the capabilities to interact with the learning analytics system through adequate user interface. Such systems enables various features such as learning recommendations, visualizations, reminders, rating and self-assessments possibilities. This paper proposes a framework for learning analytics aimed to improve personalized learning environments, encouraging the learner’s skills to monitor, adapt, and improve their own learning. It is an attempt to articulate the characterizing properties that reveals the association between learning analytics and personalized learning environment. In order to verify data analysis approaches and to determine the validity and accuracy of a learning analytics, and its corresponding to learning profiles, a case study was performed. The findings indicate that educational data for learning analytics are context specific and variables carry different meanings and can have different implications on learning success prediction.

Learning Management Systems have become fundamental in higher education for delivering and managing educational content. However, traditional implementations often lack the ability to provide personalized learning experiences and detailed insights into learner behavior. A new approach addresses these limitations by enabling more detailed Learning Analytics through the integration of interactive H5P content and the implementation of Moodle’s LogStore xAPI plugin to send Experience API-based statements within a Moodle Learning Management System. By extending this plugin, detailed user interactions, including activity outcomes, scores, durations and completion status, are captured as Learning Records and stored in a Learning Record Store for further analysis. The enriched Learning Records enable more advanced Learning Analytics that provide deeper insights into student behavior, such as identifying learning preferences, activity patterns, and knowledge levels. Future work will involve developing a recommendation system that uses the Learning Analytics data to identify the next activity best suited to fill learning gaps. The system should monitor learner preferences to maintain engagement, enable adaptive learning paths and offer personalized suggestions. Further efforts will focus on refining the system and evaluating its effectiveness in improving educational outcomes.

This research proposal is about the integration between Learning Analytics and Cognitive Computing in the development of resources to support the process of personalized learning in Higher Education and from the use of these resources verify the improvement of motivation and engagement of students in learning processes. It is a work in progress project in which is offered customized processes to students, as well as prescriptive analysis of data to teachers to support decision-making and more assertive design of learning experiences.

Higher education often underestimates the value of data in decision-making, resulting in a lack of use of big data generated from e-learning (electronic learning) activities to improve the quality of online courses and redesign teaching and learning experiences. Big data play a pivotal role in improving the quality of content and resources, accessibility, automated assessment, and understanding the impact of e-learning courses on student engagement and participation. The aim of this study was to explore students’ engagement with online learning environments to improve learners’ personalized learning experience. The research tools used were the business analytics platform “Pyramid” as a tool for analyzing student interactions within e-learning environments, and the e-learning system “Blackboard” as a tool for collecting interaction data within the e-learning environment. The data were collected through learner engagement online in the learning management system of Blackboard. A descriptive analysis approach was used to analyze the data. A total of 20,000 students from a Saudi university undertook one of the main courses during the academic year 2022--2023. Using the convenience sampling method, which is a type of nonprobability sampling strategy that allows the selection of a study sample that can be easily reached during the course of the study, the study sample of 2,600 students was selected. The results revealed significant variations in interaction rates within the Blackboard e-learning environment, along with distinct fluctuations in the duration of learner engagement on the platform. Among the four daily time periods, the morning consistently emerged as the peak time for educational activity. Mobile application access in the morning demonstrated the highest level of engagement with the course elements and content. The interaction rates gradually declined throughout the day, reaching their lowest point in the afternoon. The results also revealed that students have a high level of academic achievement. Some learning strategies have been suggested to improve students’ participation rates during low-activity periods. This study highlights the importance of using learning analytics to provide exploratory insights into learner activity and develop data-driven strategies to meet individual learner needs and improve personalized learning experiences.

The rapid advancement of artificial intelligence (AI) has revolutionized personalized learning and educational analytics, presenting new opportunities and challenges for adaptive education. This paper explores the impact of AI-driven technologies in creating personalized learning environments by examining how adaptive algorithms and data analytics shape educational experiences. The primary objective of this study is to assess the effectiveness of AI in enhancing learner engagement and outcomes through tailored instructional methods. Utilizing a mixed-method approach, this research gathers quantitative data from learning management systems to analyze engagement metrics, while qualitative insights are derived from interviews with educators and students. The findings indicate that AI-driven personalized learning significantly improves both student motivation and academic performance by adapting content to individual learning needs. Moreover, educational analytics enabled by AI offer educators critical insights into student progress, enabling proactive intervention and support. However, the study also highlights concerns regarding data privacy and the potential over-reliance on AI technologies in educational settings. These findings suggest that while AI holds transformative potential, a balanced approach is necessary to integrate technology with traditional teaching methods to ensure optimal educational outcomes. The study concludes that AI can serve as a powerful tool in enhancing personalized learning and educational analytics, provided that ethical considerations and data security are prioritized.

Learning Analytics Dashboards (LAD) are the subject of research in a multitude of schools and higher education institutions, but a lack of research into learner-facing dashboards in professional learning has been identified. This study took place in an authentic professional learning context and aims to contribute insights into LAD design by using an academic approach in a practice-based environment. An existing storytelling LAD created to support 81 accountants was evaluated using Technology Acceptance Model, finding a learner expectation for clarity, conciseness, understanding and guidance on next steps. High usage levels and a ‘take what you need’ approach was identified, with all visualizations and automated personalized feedback being considered useful although to varying degrees. Professional learners in this study focus on understanding and acting upon weaknesses rather than celebrating strengths. The lessons for LAD design are to offer choice and create elements which support learners to take action to improve performance at a multitude of time points and levels of success.

Abstract: Learning analytics have become a game-changer in education by using big data to analyse student behaviours, predict student outcomes and provide personalised interventions. This paper outlines the main components in learning analytics including data collection, predictive modelling and personalised educational strategies. It demonstrates how predictive models can be used to identify at-risk students and why real-time feedback can keep students engaged and motivated. Two case studies and examples of the data are used to illustrate how institutions can shift from reactive to proactive mode using learning analytics to track engagement, performance, and personalise the learning path. The study also shows that learning technologies are becoming adaptive to personalise learning experience, which results in a more learner-centric approach to education catering for the individual needs of students. Overall, the study demonstrates the role of learning analytics in creating a data-driven environment, which improves the student learning success and retention by addressing the challenges ahead of time.

This research study explores the conceptualization, development, and deployment of an innovative learning analytics tool, leveraging OpenAI's GPT-4 model to quantify student engagement, map learning progression, and evaluate diverse instructional strategies within an educational context. By analyzing critical data points such as students' stress levels, curiosity, confusion, agitation, topic preferences, and study methods, the tool provides a comprehensive view of the learning environment. It also employs Bloom's taxonomy to assess cognitive development based on student inquiries. In addition to technical evaluation through synthetic data, feedback from a survey of teaching faculty at the University of Iowa was collected to gauge perceived benefits and challenges. Faculty recognized the tool's potential to enhance instructional decision-making through real-time insights but expressed concerns about data security and the accuracy of AI-generated insights. The study outlines the design, implementation, and evaluation of the tool, highlighting its contributions to educational outcomes, practical integration within learning management systems, and future refinements needed to address privacy and accuracy concerns. This research underscores AI's role in shaping personalized, data-driven education.

E-Learning platforms change fast, and real-time behavioural analytics with machine learning provides the most powerful means to enhance learner outcomes. The datasets undergo preprocessing techniques like Z-score outlier detection, Min-Max scaling for feature normalization, and Ridge-RFE (Ridge regression and Recursive Feature Elimination) for feature selection in order to improve the accuracy and reliability of the predictions. Applying the Gradient Boosting Machine, classification accuracy up to a 94% level with respect to the model about predictions on learner outcomes was achievable. Thus, applying this, feedback systems may offer timely recommendations or directions in class that propel students toward better understanding on how to raise participation and success percentages. However, this approach has some potential benefits but there are still various challenges such as managing the data imbalance for models that generalize in a dynamic environment. Though hybrid methods mitigate this problem, real-time data pipelines with behaviour analytics incorporation call for significant computer-intensive resources and infrastructure. This integration has very high paybacks. It makes possible more responsive E-Learning platforms with individual needs almost met in real-time manners, thus giving instantaneous feedback, content suggestions, and timely interventions. Finally, convergence of real-time analytics with ML models culminates in adaptive learning environments which improve student engagement, retention, and quality of academic results.

There is plenty of research on the use of Big educational data, learning analytics, semantics-based knowledge modeling, and other innovative technologies within an educational environment for improving learning and tutoring. Each of these research fields is very wide and complex, and different researchers usually concentrate their work only on one or two of them and on the related subfields. The authors believe that high-quality personalized learning requires the integration of all these technologies into one intelligent educational system. Hence, the paper proposes an integrated model for combining big data learning analytics, and intelligent technologies for personalized learning and discusses the role of these technologies in achieving a more efficient learning process and interoperability between e-learning systems.

No abstract available

Adaptation through Artificial Intelligence (AI) creates individual-centered feedback strategies to reduce academic achievement disparities among students. The study evaluates the effectiveness of AI-driven adaptive feedback in mitigating these gaps by providing personalized learning support to struggling learners. A learning analytics-based evaluation was conducted on 700 undergraduate students enrolled in STEM-related courses across three different departments at Beaconhouse International College (BIC). The study employed a quasi-experimental design, where 350 students received AI-driven adaptive feedback while the control group followed traditional instructor-led feedback methods. Data were collected over 20 weeks, utilizing pre- and post-assessments, real-time engagement tracking, and survey responses. Results indicate that students receiving AI-driven adaptive feedback demonstrated a 28% improvement in conceptual mastery, compared to 14% in the control group. Additionally, student engagement increased by 35%, with a 22% reduction in cognitive overload. Analysis of interaction logs revealed that frequent engagement with AI-generated feedback led to a 40% increase in retention rates. Despite these benefits, variations in impact were observed based on prior knowledge levels and interaction consistency. The findings highlight the potential of AI-driven smart learning environments to enhance educational equity. Future research should explore long-term effects, scalability, and ethical considerations in adaptive AI-based learning systems.

Self-regulated learning (SRL) is a critical skill for lifelong learning. However, many learners struggle with SRL and need support. With the recent advancement, learning analytics (LA) has offered capabilities for supporting learning, particularly, for SRL in lifelong learning. In designing SRL analytics, recent calls asked to consider the learners' voices as well as to apply the learning theories. This study aims to explore learners' expectations of personalized support features from SRL analytics based on the learners' challenges and needs in the different phases of SRL. We conducted 10 focus group discussions with 27 students from non-formal online professional development courses. We applied thematic analysis to explore the challenges faced by learners and their expectations of SRL analytics features to support their needs. The findings highlight the importance of features to personalize goal setting, progress tracking, socio-emotional and motivational support, and feedback among other features, in facilitating SRL. The results of the study provide insights into the design of SRL analytics that can effectively support learners.

AI-powered educational technology that is designed to support teachers in providing personalized instruction can enhance their ability to address the needs of individual students, hopefully leading to better learning gains. This paper presents results from a participatory research aimed at co-designing with science teachers a learning analytics tool that will assist them in implementing a personalized pedagogy in blended learning contexts. The development process included three stages. In the first, we interviewed a group of teachers to identify where and how personalized instruction may be integrated into their teaching practices. This yielded a clustering-based personalization strategy. Next, we designed a mock-up of a learning analytics tool that supports this strategy and worked with another group of teachers to define an ‘explainable learning analytics’ scheme that explains each cluster in a way that is both pedagogically meaningful and can be generated automatically. Third, we developed an AI algorithm that supports this ‘explainable clusters’ pedagogy and conducted a controlled experiment that evaluated its contribution to teachers’ ability to plan personalized learning sequences. The planned sequences were evaluated in a blinded fashion by an expert, and the results demonstrated that the experimental group – teachers who received the clusters with the explanations – designed sequences that addressed the difficulties exhibited by different groups of students better than those designed by teachers who received the clusters without explanations. The main contribution of this study is twofold. First, it presents an effective personalization approach that fits blended learning in the science classroom, which combines a real-time clustering algorithm with an explainable-AI scheme that can automatically build pedagogically meaningful explanations from item-level meta-data (Q Matrix). Second, it demonstrates how such an end-to-end learning analytics solution can be built with teachers through a co-design process and highlights the types of knowledge that teachers add to system-provided analytics in order to apply them to their local context. As a practical contribution, this process informed the design of a new learning analytics tool that was integrated into a free online learning platform that is being used by more than 1000 science teachers.

No abstract available

Higher education still follows a traditional one-size-fits-all, centralized, controlled, and static learning/teaching approach. Moreover, there is a mismatch between the competencies acquired by the student and the industry needs. Addressing the problem should include effective cooperation with industry, better integration with innovative research, and internationalization. We require a move away toward a more personalized, networked, agile, and industry-oriented model for learning. In this research work, we propose a Scenario for using learning analytics to recommend personalized micro-credentials based on a competences score as a solution for reducing the gap between the industry needs and the actual students’ skill set.

Conventional approaches such as Logistic Regression and Decision Trees have difficulties in accurately forecasting student progress in higher education due to their limited capacity to capture intricate data linkages. This study introduces a new method that utilizes a Graph Attention Network (GAT) to represent complex patterns and relationships in student data. The procedure involves data cleansing, standardization, natural language processing, feature extraction, and graph generation. The GAT model demonstrated exceptional performance, surpassing standard models, with an accuracy of 0.97, precision of 0.96, recall of 0.98, F1-Score of 0.979, and AUC-ROC of 0.99. The results illustrate the GAT model’s resilience and efficacy, providing a powerful instrument for educational analytics to increase personalized learning interventions and better student outcomes.

Artificial Intelligence (AI) has revolutionized education by enabling personalized learning experiences through adaptive platforms. However, traditional AI-driven systems primarily rely on correlation-based analytics, limiting their ability to uncover the causal mechanisms behind learning outcomes. This study explores the integration of Knowledge Graphs (KGs) and Causal Inference (CI) as a novel approach to enhance AI-driven educational systems. KGs provide a structured representation of educational knowledge, facilitating intelligent content recommendations and adaptive learning pathways, while CI enables AI systems to move beyond pattern recognition to identify cause-and-effect relationships in student learning. By combining these methods, this research aims to optimize personalized learning path recommendations, improve educational decision-making, and ensure AI-driven interventions are both data-informed and causally validated. Case studies from real-world applications, including intelligent tutoring systems and MOOC platforms, illustrate the practical impact of this approach. The findings contribute to advancing AI-driven education by fostering a balance between knowledge modeling, adaptability, and empirical rigor.

The integration of artificial intelligence (AI) in education holds significant promise for transforming personalized learning. By analyzing student learning data, AI systems can adapt instruction to meet individual needs through tailored content, adaptive learning paths, real-time feedback, and continuous improvement loops. However, effective personalization at scale demands not only access to large volumes of learner data but also robust data architectures to collect, organize, standardize, and analyze that data in a secure and meaningful way. However, note that the ability of AI to personalize learning requires data about the learner and prior learning. Personalization at scale requires data at scale. The Architecture for AI-Augmented Learning (A4L) frame-work addresses these needs by establishing a comprehensive data pipeline that supports AI-driven personalization. This pipeline introduced capabilities for direct data ingestion, anonymization, and standardization, as well as integrated analytics and visualization pipelines to deliver actionable insights to educators and learners alike.

No abstract available

The proliferation of online higher education has underscored the need for innovative approaches to enhance student learning, engagement, and success. This paper explores the transformative potential of artificial intelligence (AI) in revolutionizing online education. By focusing on personalized learning, AI-driven assessment, and student engagement, this research investigates how AI technologies can create tailored educational experiences, optimize learning outcomes, and foster a dynamic online learning environment. The study delves into the implementation of AI-powered tools, such as intelligent tutoring systems, adaptive learning platforms, and predictive analytics, to address individual student needs, provide timely feedback, and promote active participation. Through a comprehensive analysis of the existing literature and emerging trends, this paper aims to identify key challenges, opportunities, and best practices for leveraging AI to optimize online higher education, ultimately contributing to improved student satisfaction, retention, and academic achievement.

No abstract available

Despite the globalization of educational content, language remains a significant barrier. When translating educational content, multilingual translation has become crucial to meet this challenge, with an emphasis on incorporating the cultural context of the target country and the educational context of the learners. However, existing machine translation systems often fail to adequately account for these contextual factors. This study explores the potential of the Large Language Model(LLM) to improve the translation of assessment items through In-context Learning. Two prompt engineering strategies are compared: the ‘assessment-aware prompt’, which includes only the specifications of the assessment, and the ‘curriculum-aware prompt’, which includes the educational and cultural context of the target country in addition to the assessment specifications. From the comparison of linguistic features and the expert reviews, we found that the curriculum-aware translation produced more valid and feasible results, highlighting the effectiveness of LLM-based automatic translation methods that integrate curriculum context.

Background Story-centered curricula (SCC) can effectively enhance clinical reasoning and learner engagement; however, developing high-fidelity scenarios is resource-intensive. We developed an SCC using a large language model (LLM; ChatGPT, GPT-4) to generate longitudinal respiratory cases for physiotherapy students and evaluated its educational impact. Methodology This single-institution, quasi-experimental study used a non-randomized historical control design. Physiotherapy students in a 15-session course were divided into an SCC group, which completed eight LLM-generated narrative sessions, and a control group, which received traditional case-based sessions without a continuous storyline. The primary outcome was the change in score on a 30-item, 300-point proficiency test administered before and after the intervention. Secondary outcomes included five Likert-scale items evaluating learner experience. Group comparisons used appropriate parametric or non-parametric tests, and multivariable linear regression adjusted for age, sex, program, and pre-test score. Results The final sample consisted of 169 participants, with 92 in the SCC group and 77 in the control group. Baseline characteristics and pre-test median scores were 90 with an interquartile range (IQR) of 70 to 100 in the SCC group and 90 with an IQR of 70 to 110 in the control group, showing no significant differences between groups. The SCC group demonstrated a greater improvement in test performance, with a median change of 105 and an IQR of 78 to 140, compared with a median change of 80 and an IQR of 50 to 110 in the control group (p < 0.001). The SCC group also achieved higher post-test scores, with a median of 195 and an IQR of 160 to 223, compared with a median of 180 and an IQR of 150 to 200 in the control group (p = 0.013). Positive questionnaire responses (scores of 4 or 5) exceeded 90% across all domains, including immersion 87 (94.6%) and learning retention 88 (95.7%). Participation in the SCC program remained an independent predictor of post-test performance, with a regression coefficient of 23.87 and a 95% confidence interval of 11.32 to 36.42 (p < 0.001). Conclusions An SCC utilizing an LLM is an innovative educational approach that effectively balances improved learning outcomes with efficient scenario development, offering significant potential to advance simulation-based education in physiotherapy.

This paper presents a human-computer collaborative method for constructing a curriculum knowledge system to improve the low efficiency due to heavy reliance on manual annotation for knowledge graph construction in traditional course. This method first manually extracts the “chapter-section-knowledge unit” structure from the textbook to construct an initial knowledge tree, and then employs retrieval-augmented techniques, with a large language model, to generate high-quality embeddings for each knowledge point by integrating relevant contextual information. Finally, a knowledge graph completion algorithm is applied to uncover horizontal and cross-chapter relationships among knowledge points through semantic computation, completing the transformation from a hierarchical knowledge tree to a fully connected knowledge graph. This method retains the interpretability and scalability inherent to knowledge graphs with no need for large-scale data annotation. The proposed method enables the efficient construction of a multimodal, fine-grained curriculum knowledge system, thereby providing strong support for tailored teaching.

No abstract available

Student reliance on Large Language Model systems like ChatGPT, to assist with their assignments, has become the norm over the last few years. Educators have applied different strategies to deal with their use. One common method is to threaten academic misconduct if submitted work is not authentic. Another technique uses a rigorous assessment model that ensures students meet the curricular outcomes of the course. To demand students avoid using these modern tools seems impractical and deprives them of an important resource that can assist in their learning. Although a thorough assessment model may give guidance to students in how to properly use these systems, as it reinforces that they must know the material they submit, it ignores how these systems could help students develop a more in-depth understanding of the material if integrated directly into the curriculum. In this preliminary work we look at how the use of non-trivial assignments could benefit from encouraging students to use these systems and how to potentially scaffold the work to create an optimal learning experience.

The rapid advancement of Large Language Models (LLMs) has revolutionized personalized learning by enabling precise content generation and real-time interaction. However, integrating LLMs into learning pipelines poses challenges such as latency and state management. This paper presents PathFinder, a low-latency adaptive assessment system that uses an LLM-driven engine for real-time curriculum development. Students select topics and assess their proficiency through adaptive multiple-choice questions, with difficulty adjusting dynamically. User selections and ratings are cached in Redis for fast access, while responses and evaluations are stored in MongoDB for analysis. Built on a MERN (MongoDB–Express–React–Node) architecture, PathFinder leverages four LLM-powered services to generate quizzes, identify skill gaps, prioritize modules, and provide relevant resources–delivering a personalized and dynamic learning experience.

Foundational computer science courses, epitomized by Discrete Mathematics, serve as cornerstones for students’ academic progression in the field. However, these essential subjects frequently present considerable hurdles in both teaching and learning. This paper delves into the transformative potential of Large Language Models (LLMs) in navigating these complexities and instigating meaningful pedagogical reform. Through a focused case study on the Discrete Mathematics curriculum, we meticulously investigate the multifaceted integration of LLMs across diverse instructional dimensions. This includes leveraging LLMs for the efficient generation of course materials, the facilitation of personalized learning pathways tailored to individual student needs, the creation of engaging and interactive exercises that foster deeper understanding, and the development of innovative assessment strategies. Our exploration extends to a thorough discussion of the salient advantages offered by this LLM-driven approach, while also acknowledging its inherent limitations and charting promising avenues for future research and implementation in this crucial educational domain.

No abstract available

Background Disparities in hematology care affect people living with sickle cell disease, hemophilia in females (XX), and Duffy null antigen status, to name a few. Integrating health equity content into the first-year medical student (MS1) preclinical curriculum is crucial for training physicians to provide equitable, comprehensive care. The preclinical years present a unique opportunity wherein essential topics and a framework for clinical reasoning are taught prior to rotating on medical services. This makes it prime for incorporating health equity content that will later inform medical care. The Yale Health Equity Thread within the Yale School of Medicine (YSM) has defined several domains of health equity, including (1) race & ethnicity, (2) sex & gender, and (3) sexual orientation & gender identity. YSM faculty were previously queried regarding barriers to incorporating health equity content into the MS1 curricula and noted time to develop content, an already packed curriculum, and lack of expertise with this content as barriers. The hematology MS1 course includes lectures and small group workshops that center on clinical cases. Here we examine the use of a large language model (LLM), Humata.ai to identify gaps in health equity domains and address barriers to incorporating health equity content into the MS1 hematology workshop. Methods During the 2023- 2024 academic year, 11 instructors led six small group workshops on Bleeding Disorders and Thrombosis. The workshop was provided to Humata.ai. Humata.ai was prompted to identify gaps in the three health equity domains. An audit summary was generated for each clinical case in the workshop and presented to instructors for review. Instructors were then invited to complete a survey and free text questions to assess their desire to incorporate health equity content into the course, identify barriers, and rate the utility of the audit in improving health equity content into the workshops. Results There were 7 (64%) survey responses. All respondents expressed a desire to include sex & gender, and race & ethnicity content. There were no reported barriers to incorporating sex & gender or race & ethnicity content. For sexual orientation & gender identity, 72% said “yes” and 28% “maybe” regarding desire to incorporate content. Barriers to incorporating sexual orientation & gender identity content included unfamiliarity with the topic (28%) and lack of time to update course material (28%). Sex & gender content: 43% of respondents selected “yes” and 57% “maybe” to the Humata.ai audit being helpful with the content. 100% plan to incorporate the suggestions into the small group workshops. Sexual orientation & gender identity: 57% of respondents selected “yes” and 43% “maybe” to the Humata.ai audit being helpful with the content. 71% plan to incorporate the suggestions, while 43% “maybe.” Race & ethnicity: 29% of respondents selected “yes”, 57% “maybe” and 14% “no.” 57% plan to incorporate the suggestions, and 43% “maybe.” Qualitative responses: “The sexual orientation and gender identity prompts were the most helpful followed by the sex and gender-based prompts.” “I think that the audits could be particularly helpful in terms of discussing gender-specific patient presentations and inclusion of teaching material for transgender and non-binary individuals. Although admittedly, I would have probably discussed many of the suggestions if prompted to without the assistance of the AI audit.” “Discussion of gender affirming hormone therapy and thrombosis [was helpful].” Conclusion: Our study demonstrates the potential of leveraging LLMs to enhance inclusion of health equity content into the MS1 curricula. Despite the small number of respondents, the responses signal a strong preference to incorporate health equity content. Barriers to sexual orientation & gender identity inclusion were due to unfamiliarity with the topic and time constraints. Humata.ai audit was helpful in addressing sex & gender-based prompts, but not so much race & ethnicity. The use of LLM to audit course material might decrease barriers to incorporating such content by prompting instructors to consider gaps in curricula. Future steps include ongoing survey of workshop leaders for the upcoming academic year and expanding the audit to include lectures and other workshops and team-based learning activities.

: With the rapid advancement of artificial intelligence, Large Language Models (LLMs) have become a critical technology, significantly transforming the educational landscape. This paper investigates the application of LLMs in the "Innovation and Entrepreneurship" course, emphasizing their potential to revolutionize course design and implementation. By leveraging LLMs, educators can innovate teaching content, personalize learning experiences, and effectively integrate interdisciplinary knowledge. This study explores the use of LLMs for personalized tutoring, creative proposal generation, real-time feedback, intelligent case analysis, and adaptive learning resource recommendations. The findings indicate that LLMs can enhance the richness and practicality of teaching materials, improve the personalization and interactivity of the learning experience, and ultimately foster a more intelligent and efficient learning environment for future entrepreneurs.

In recent years, methods based on pre-trained large models have received extensive attention in the field of natural language processing. The large language model has the potential to reduce the development cost of question answering systems in the education field and improve the accuracy of question answering system. In this paper, based on the curriculum standards of compulsory education, an educational knowledge base is constructed by using the method of label alignment optimization, and a professional large language model question answering system is constructed based on the large language model to meet the requirements of the new curriculum standards, so as to generate a high-availability teaching scheme. Secondly, a prompt word method based on educational keyword template is proposed, aiming at improving the quality of question answering generation of large language model in educational question answering system. The experimental results show that this system is superior to the universal large language model in both objective and subjective evaluation, and has higher usability in the field of education and more accuracy in generating answers.

Primary mathematics education faces systemic challenges in translating curriculum reforms into classroom practice, exacerbated by teachers’ cognitive overload and limited support for pedagogical innovation. This study develops an Intelligent Teaching Design Assistant grounded in socio-constructivist and cognitive load theories to address these challenges. Thirty-four primary mathematics teachers participated in a quasi-experimental study. The Intelligent Teaching Design Assistant integrates Large Language Models with multi-dimensional knowledge bases (curriculum standards, teaching strategies, student profiles) and a multi-agent architecture (process planner, student simulator). The Intelligent Teaching Design Assistant significantly outperformed generic Large Language Models, improving overall lesson plan quality. This work pioneers a replicable pathway for AI to empower teacher agency and advance 21st-century educational transformation.

The integration of Generative Artificial Intelligence (GenAI) into the development of theatre/drama curricula offers a multitude of opportunities, and simultaneously presents complex challenges. This article explores the potential benefits and complications of utilizing AI in the development of theatre/drama curriculum, emphasizing the critical need for innovative research practices to maximize AI's effectiveness in pedagogy. I assert that ongoing exploration and research is imperative to identify best practices in the nascent relationship between AI and educators. I describe the efforts of BYU Theatre Education faculty to explore AI-driven content curation and recommendation algorithms as a means of strengthening curriculum design. I assert that this can only happen when Artificial Intelligence directives are paired with informed, embodied, decision-making processes that preserve the richness of drama pedagogies.

No abstract available

This study proposes and empirically validates a large-language-model(LLM) driven framework that aligns undergraduate majors and course portfolios with the competency requirements of specific job roles. Using Samsung Electronics’ Device Solutions division as a testbed, we analyzed two highly representative semiconductor positions System LSI circuit-architecture design and R&D Center process-architecture integration because they respectively demand the narrowest and broadest disciplinary scopes in the industry. The 4o LLM was sequentially prompted with structured job descriptions, 6 STEM majors at KAIST, and 248 associated courses. A two-tier evaluation quantified major-level (Level 1) and course-level (Level 2) relevance both before and after Knowledge Injection (KI) of the “Recommended Subject” field in each job description. KI reinforced Electrical Engineering’s dominance in circuit design by + 1.1% total score but markedly broadened multidisciplinary suitability for process integration by + 7.7%. As a result, Mechanical Engineering and Chemistry were elevated into a higher relevance tier according to the evaluation criteria. At the course level, all Recommended Subjects and their advanced courses consistently ranked among the highest within each major, and they were found to be most susceptible to the effects of KI. This confirms their utility as anchors for automated job education mapping. The results demonstrate that KI granularity should mirror role characteristics core-focused for design-centric jobs and dispersive for integration-oriented jobs. Beyond extending LLM use to curriculum design and hiring-fit assessment, the framework offers a scalable pathway for universities, firms, and policymakers to co-engineer evidence-based talent pipelines and mitigate looming bachelor-level workforce shortages in semiconductors.

This study investigates the application effectiveness of the Large Language Model (LLMs) ChatGLM in the automated generation of high school information technology exam questions. Through meticulously designed prompt engineering strategies, the model is guided to generate diverse questions, which are then comprehensively evaluated by domain experts. The evaluation dimensions include the Hitting(the degree of alignment with teaching content), Fitting (the degree of embodiment of core competencies), Clarity (the explicitness of question descriptions), and Willing to use (the teacher's willingness to use the question in teaching). The results indicate that ChatGLM outperforms human-generated questions in terms of clarity and teachers' willingness to use, although there is no significant difference in hit rate and fit. This finding suggests that ChatGLM has the potential to enhance the efficiency of question generation and alleviate the burden on teachers, providing a new perspective for the future development of educational assessment systems. Future research could explore further optimizations to the ChatGLM model to maintain high fit and hit rates while improving the clarity of questions and teachers' willingness to use them.

Despite well‐designed curriculum materials, teachers often face challenges implementing them due to diverse classroom needs. This paper investigates whether large language models (LLMs) can support middle school math teachers by helping create high‐quality curriculum scaffolds, which we define as the adaptations and supplements teachers employ to ensure all students can access and engage with the curriculum. Through cognitive task analysis with expert teachers, we identify a three‐stage process for curriculum scaffolding: observation, strategy formulation and implementation. We incorporate these insights into three LLM approaches to create warmup tasks that activate students' background knowledge. The best‐performing approach provides the model with the original curriculum materials and an expert‐informed prompt; this approach generates warmups that are rated significantly higher than those created by expert teachers in terms of alignment to learning objectives, accessibility to students working below grade level and teacher preference. This research demonstrates the potential of LLMs to support teachers in creating effective scaffolds and provides a methodology for developing artificial intelligence‐driven educational tools. What is already known about this topic Scaffolding is essential for enabling students to access and engage with curriculum materials. Large language models (LLMs) have shown promise in generating educational content and supporting teachers. Teachers frequently need to adapt and supplement standardized curricula to meet the diverse needs of their students. What this paper adds Identifies a three‐stage curriculum scaffolding process (observation, strategy formulation, implementation) used by expert teachers. Demonstrates that providing LLMs with additional context from the curriculum, such as the original warmup task, helps to ground the model and improve the quality of the generated warmup tasks. Demonstrates that, when prompted well, LLMs can generate warmup tasks that are of similar or better quality compared to those created by expert teachers in terms of alignment to learning objectives, accessibility and teacher preference. Implications for practice and/or policy Provides practical suggestions for prompting LLMs to generate high‐quality warmup tasks for middle school math teachers, such as incorporating additional curriculum context and expert‐informed prompts. Demonstrates how cognitive task analysis with expert teachers can be used to develop LLM‐based tools for educators that align with their practices and preferences. Indicates that additional research is needed to explore the potential for LLMs to support other types of curriculum adaptations, evaluate their effectiveness in classroom settings and investigate how they can be effectively tailored to the specific needs and characteristics of individual students.

The alignment of learning materials with the learning objectives (LOs) is critical for successfully implementing the Problem-Based Learning (PBL) curriculum. This study investigated the capabilities of Gemini Advanced, a large language model (LLM), in creating clinical vignettes that align with LOs, and comprehensive tutor guides. This study used a faculty-written clinical vignette about diabetes mellitus for third-year medical students. We submitted the LOs and the associated clinical vignette and tutor guide to the LLM to evaluate their alignment and generate new versions. Four faculty members compared both versions using a structured questionnaire. The mean evaluation scores for original and LLM-generated versions are reported. The LLM identified new triggers for the clinical vignette to align it better with the LOs. Moreover, it restructured the tutor guide for better organization and flow and included thought-provoking questions. The medical information provided by the LLM was scientifically appropriate and accurate. LLM-generated clinical vignette scored higher (3.0 vs. 1.25) for the alignment with the LOs. However, the original version scored better for being educational-level appropriate (2.25 vs. 1.25) and adhering to PBL design (2.50 vs. 1.25). The LLM-generated tutor guide scored higher for better flow (3.0 vs. 1.25), comprehensive and relevant content (2.75 vs. 1.50) and thought-provoking questions (2.25 vs. 1.75). However, LLM-generated learning material lacked visual elements. In conclusion, this study demonstrated that Gemini could align and improve PBL learning materials. By leveraging the potential of LLMs while acknowledging their limitations, medical educators can create innovative and effective learning experiences for future physicians.

Effective lesson planning is crucial in education process, serving as the cornerstone for high-quality teaching and the cultivation of a conducive learning atmosphere. This study investigates how large language models (LLMs) can enhance teacher preparation by incorporating them with Gagne's Nine Events of Instruction, especially in the field of mathematics education in compulsory education. It investigates two distinct methodologies: the development of Chain of Thought (CoT) prompts to direct LLMs in generating content that aligns with instructional events, and the application of fine-tuning approaches like Low-Rank Adaptation (LoRA) to enhance model performance. This research starts with creating a comprehensive dataset based on math curriculum standards and Gagne's instructional events. The first method involves crafting CoT-optimized prompts to generate detailed, logically coherent responses from LLMs, improving their ability to create educationally relevant content. The second method uses specialized datasets to fine-tune open-source models, enhancing their educational content generation and analysis capabilities. This study contributes to the evolving dialogue on the integration of AI in education, illustrating innovative strategies for leveraging LLMs to bolster teaching and learning processes.

Conventional teacher assessment is labor-intensive and subjective. Prior LLM-based systems improve scale but rely on post hoc rationales and lack built-in trust controls. We propose an explainable-by-design framework that couples (i) Dual-Lens Hierarchical Attention—a global lens aligned to curriculum standards and a local lens aligned to subject-specific rubrics—with (ii) a Trust-Gated Inference module that combines Monte-Carlo-dropout calibration and adversarial debiasing, and (iii) an On-the-Spot Explanation generator that shares the same fused representation and predicted score used for decision making. Thus, explanations are decision-consistent and curriculum-anchored rather than retrofitted. On TeacherEval-2023, EdNet-Math, and MM-TBA, our model attains an Inter-Rater Consistency of 82.4%, Explanation Credibility of 0.78, Fairness Gap of 1.8%, and Expected Calibration Error of 0.032. Faithfulness is verified via attention-to-rubric alignment (78%) and counterfactual deletion tests, while trust gating reduces confidently wrong outputs and triggers reject-and-refer when uncertainty is high. The system retains 99.6% accuracy under cross-domain transfer and degrades only 4.1% with 15% ASR noise, reducing human review workload by 41%. This establishes a reproducible path to trustworthy and pedagogy-aligned LLMs for high-stakes educational evaluation.

This study investigates how generative AI models (ChatGPT, Claude and Gemini) can be systematically integrated into curriculum design using Hilda Taba’s inductive model. Addressing Sustainable Development Goal 4 (SDG 4), it introduces the EduCompass framework to enhance inclusivity and instructional quality. A structured needs assessment with MBA students identified skill gaps in analytics and AI ethics. Deep research prompts were fed into ChatGPT, Claude and Gemini to generate curriculum components. Outputs were evaluated on accuracy, relevance and clarity by academic experts and analyzed for readability (Flesch–Kincaid), semantic similarity (cosine) and statistical significance (ANOVA). ChatGPT produced the most readable and pedagogically sound content, followed by Gemini and Claude. ChatGPT also received the highest expert ratings for clarity, accuracy and relevance, with significant differences confirmed via ANOVA. Cosine similarity showed the highest conceptual overlap between ChatGPT and Gemini. This is a study to empirically compare AI-generated curricula across multiple LLMs within a classical curriculum framework. EduCompass offers a replicable model for AI-enhanced, SDG-aligned curriculum design in higher education.

. External learning, occurring outside traditional classrooms, emphasizes learner autonomy, flexibility, lifelong learning, critical thinking, and networking. However, it also presents challenges such as limited structure, resource access issues, isolation, distractions, progress assessment difficulties, credibility recognition gaps, and technology barriers exacerbated by the digital divide. With new technologies and innovative methods constantly emerging in the world of education, one such recent development is the use of large language models as a facilitator of learning outside the classroom. To enhance the external learning experience, a workflow incorporating AI tools like Language Model Large (LLM) is proposed. This workflow spans goal setting, curriculum curation, interactive learning sessions, and progress tracking, leveraging LLM capabilities to personalize the learning journey, curate relevant resources, facilitate interactive engagement, and provide real-time feedback. This paper examines the utilization of LLM by presenting the application of LLM in different external learning scenarios.

Artificial intelligence (AI) has permeated all human activities, bringing about significant changes and creating new scientific and ethical challenges. The field of education could not be an exception to this development. OpenAI’s unveiling of ChatGPT, their large language model (LLM), has sparked significant interest in the potential applications of this technology in education. This paper aims to contribute to the ongoing discussion on the role of AI in education and its potential implications for the future of learning by exploring how LLMs could be utilized in the teaching of mathematics in higher education and how they compare to the currently widely used computer algebra systems (CAS) and other mathematical tools. It argues that these innovative tools have the potential to provide functional and pedagogical opportunities that may influence changes in curriculum and assessment approaches.

In the context of the evolving digital era, this study introduces a conceptual model that integrates Education for Sustainable Development (ESD) into English for Academic Purposes (EAP) writing, leveraging Critical Digital Literacies (CDL). It aims to develop EFL doctoral students' competencies necessary for sustainable development, addressing the gap in current genre-based EAP instruction that often neglects comprehensive ESD competencies such as systems thinking and self-awareness. Utilizing innovative tools like ChatGPT within a CDL framework, the model focuses on both explicit teaching objectives related to genre knowledge and implicit ones tied to ESD. Specifically, the study outlines eight scenarios where GPT technology aids in achieving these pedagogical goals. By implementing this model, we aim not only to enrich the EAP writing curriculum but also to subtly shift educational practices towards embracing the values of sustainability and digitalization.

PURPOSE OF REVIEW Large language model (LLM) use is rapidly growing among trainees and educators in clinical medicine, and new developments with LLM-augmented educational approaches have the potential to enhance learning and ease educator burden. We review LLM fundamentals, their use in clinical education, and how they may be integrated into trauma anesthesia education. RECENT FINDINGS Traditional trauma anesthesia curricula and learning objectives lack granularity and the ability to meet unique learner needs. New approaches to clinical education and curriculum development are leveraging LLM capabilities, but applications for trauma anesthesia education remain insufficiently evaluated. SUMMARY Augmenting traditional trauma anesthesia educational approaches with LLMs bears promise. Integrating numerous trauma anesthesia curricula and enhancing educational tools with LLMs to evaluate learner performance and satisfaction and educator burden would be informative to help refine best practices and target areas for improvement.

This poster presents a methodology for evaluating course syllabi using large language models and semantic embeddings. We demonstrate this approach by analyzing 141 cybersecurity syllabi from top-ranked American institutions. Using the intfloat/e5-base-v2 embedding model, syllabus text was mapped to competencies in the National Institute of Standards and Technology's (NIST) National Initiative for Cybersecurity Education Framework (NICE Framework). Cybersecurity was chosen as the domain due to the field's workforce shortage and the existence of the NICE Framework as a standardized taxonomy. This research illustrates how large language models can support curriculum analysis by automating the alignment of educational materials to competency frameworks, offering a scalable approach for other disciplines.

This study verifies the ability of large language models (LLMs) to generate a curriculum and develop syllabi for specific courses. We prompted four models to generate two sets of curricula for a bachelor’s degree in Economics and Management. We also generated syllabi for the courses included in the curriculum. We chose five Polish public economics universities offering those degree programs for comparison. Four LLMs were used in this experiment: ChatGPT-3.5, ChatGPT-4, Google Bard, and Gemini. Two of them are multimodal models. The study used an iterative approach, increasing the detail of the prompt in each iteration. The results show that the more specific prompt is given to the LLM, the less accurate the results are. Moreover, the experiment shows that none of the LLMs developed a complete curriculum at a level comparable to that generated by humans. However, LLMs can significantly help create a curriculum and develop syllabi by humans, provided that there is close human–artificial intelligence (AI) collaboration. The results obtained from the AI-assisted curriculum design differ depending on the model. By analyzing the differences between the tools and the real degree programs and syllabi, we determined that multimodal models are better suited for this task than older models.

With the rapid development of AI technology, personalized learning path planning is becoming an important research field. Dual system teaching models have great influence on the learning effect of students. Therefore, how to optimize the learning process of students through the personalized learning path is a great challenge. A Deep Reinforcement Learning (DRL) algorithm for personalized learning path planning is proposed. Based on deep Q-networks and reinforcement learning feedback mechanisms, the algorithm achieves adaptive optimization. The experimental results show that the algorithm based on the DRL is superior to the traditional method in learning performance, learning efficiency and satisfaction. Compared with traditional path planning algorithms based on rules and collaborative filtering, the proposed algorithm improves learning performance by 20%. Moreover, it has obvious advantages in learning motivation and efficiency. This paper provides a new way of thinking to intelligent education system, especially dual-system teaching model, and it is of great practical significance.

The rapid growth of artificial intelligence (AI) has significantly reshaped the landscape of education, particularly in vocational English training, which demands personalized and adaptive learning approaches. As a core component of Smart Education, AI-driven frameworks offer opportunities to transform traditional teaching methods into intelligent, data-driven, and learner-centered systems. This study proposes an innovative adaptive learning path recommendation framework based on deep learning, aiming to address challenges such as learner diversity, engagement, and efficiency. This study proposes an innovative adaptive learning path recommendation framework based on deep learning, aiming to address challenges such as learner diversity, engagement, and efficiency. By integrating attention mechanisms and collaborative filtering, the framework dynamically captures contextual dependencies and user-specific behaviors to generate personalized learning paths. The methodology involves preprocessing learner interaction data, embedding it into a high-dimensional space, and utilizing a deep neural network enhanced with attention mechanisms to predict resource relevance. Furthermore, the integration of optical technologies, such as eye-tracking and virtual reality (VR), provides a novel approach to capturing real-time learner engagement and interaction patterns. The framework was validated through experiments on a real-world vocational English dataset, involving over 5,000 learners. Results demonstrate that the proposed model outperforms traditional collaborative filtering and non-attention-based deep learning models, achieving superior precision, recall, and F1-Score. Additionally, learners following the recommended paths exhibited a 25% faster course completion rate and a 30% higher retention rate compared to static, non-adaptive approaches. These findings highlight the framework's potential to enhance learning outcomes and engagement in vocational English education. This work bridges the gap between AI-driven recommender systems, optical technologies, and practical educational applications, offering a scalable, dynamic, and effective solution for personalized learning in diverse educational contexts.

The implementation of Outcome-Based Education (OBE) in engineering courses poses practical challenges for supporting diverse learners as they navigate complex competency structures. While traditional teaching often follows a linear path, many students benefit from more adaptive, personalized guidance. To explore how AI might support this process, we developed OBE-Navigator, a multi-agent platform built on the Coze framework and delivered via WeChat Mini Program. The system includes three role-differentiated agents: an AI Tutor Expert, a teacher assistant, and a data analyst. We conducted a mixed-methods usability study with 14 undergraduate students enrolled in a microcontroller course, using think-aloud sessions, System Usability Scale (SUS) questionnaires, and semi-structured interviews. The average SUS score was 74.8 (SD = 9.5), indicating a generally positive user experience. Qualitative analysis revealed that students found value in the learning path recommendations and instant feedback, though some encountered confusion when switching between agents. These findings suggest that multi-agent systems (MAS) can play a supportive role in OBE-aligned learning, especially when attention is paid to interface clarity and user flow.

In the digital era, the integration of technology into classrooms has transformed traditional learning environments. Among these advancements, Intelligent Tutoring Systems (ITS) stand out as one of the most promising innovations in educational technology, offering personalized and adaptive learning opportunities. This study focuses on developing a College English Intelligent Tutoring System with efficient learning path planning using Artificial Intelligence (AI). The research introduces a novel Intelligent Honeybee Mating Optimized Dynamic Long Short-Term Memory (IHMO-DLSTM) model to enhance recommendation accuracy and efficiency. Inspired by honeybee mating behaviors, the IHMO method effectively balances exploration and exploitation in optimization tasks. The DLSTM component analyzes students’ learning behaviors and generates personalized recommendations based on their unique learning paths. The system leverages data on student performance, behaviors, preferences and interactions to create a foundation for tailored learning pathways. Data preprocessing involves cleaning and normalization to ensure the quality and reliability of the dataset. The proposed IHMO-DLSTM technique addresses the challenge of low recommendation accuracy in personalized learning and significantly improves the efficiency of the recommendation process. Experimental results demonstrate that the IHMO-DLSTM model outperforms existing approaches, achieving a precision of 0.9122, a recall of 0.9345 and an F1-score of 0.9560. These metrics indicate more accurate, comprehensive, and customized learning path recommendations. The results also confirm that the proposed method substantially improves alignment between personalized learning paths and learner needs, delivering better outcomes compared to traditional algorithms. The findings highlight the potential of the proposed College English Intelligent Tutoring System to revolutionize personalized education by providing accurate and efficient learning path recommendations, enhancing both student engagement and educational outcomes.

This paper focuses on the relevant research on the application of artificial intelligence (AI) technology in the teaching of basic computer courses in universities. Traditional learning platforms have problems such as poor resource integration and lack of personalized learning support. Based on this, an intelligent learning path architecture relying on AI has been constructed. The construction basis and functional characteristics of this architecture are elaborated, and practical strategies are proposed from the aspects of knowledge graph construction and intelligent resource recommendation. Verified by practical application, this path can effectively enhance learning efficiency and teaching quality, and can provide a reference for the teaching reform of basic computer courses in colleges and universities.

With the in-depth advancement of smart education, personalized education has emerged as a significant trend in educational development. Traditional teaching models and tools struggle to meet the diverse needs arising from students' individual differences, while the integration of knowledge graphs and artificial intelligence (AI) offers a new solution for personalized educational services. In this context, an adaptive learning path planning system is proposed based on the integration of knowledge graphs and AI. Specifically, the system constructs a three-dimensional knowledge graph through the major courses to clearly show the dependency and hierarchical structure between knowledge nodes between courses. Meanwhile, a personalized decision-making module is integrated to dynamically collect learners' learning information and make planning decisions. Moreover, an intelligent interaction system is constructed based on the knowledge graph. The experimental results obtained from the evaluation systems using the open-source tools Lighthouse and Coze platforms demonstrate the stability and effectiveness of the proposed system. This system offers novel approaches to addressing the disconnect between what learners learn and what they need, while also providing a replicable technical solution for the large-scale promotion of personalized education.

E-learning courses often suffer from high dropout rates and low student satisfaction. One way to address this issue is to use personalized learning paths (PLPs), which are sequences of learning materials that meet the individual needs of students. However, creating PLPs is difficult and often involves combining knowledge graphs (KGs), student profiles, and learning materials. Researchers typically assume that the problem of creating PLPs belong to the nondeterministic polynomial (NP)-hard class of computational problems. However, previous research in this field has neither defined the different variations of the PLP problem nor formally established their computational complexity. Without clear definitions of the PLP variations, researchers risk making invalid comparisons and conclusions when they use different metaheuristics for different PLP problems. To unify this conversation, this article formally proves the NP-completeness of two common PLP variations and their generalizations and uses them to categorize recent research in the PLP field. It then presents an instance of the PLP problem using real-world data and shows how this instance can be cast into two different NP-complete variations. This article then presents three artificial intelligence (AI) strategies, solving one of the PLP variations with back-tracking and branch and bound heuristics and also converting the PLP variation instance to XCSP${}^{3}$, an intermediate constraint satisfaction language to be resolved with a general constraint optimization solver. This article solves the other PLP variation instance using a greedy search heuristic. The article finishes by comparing the results of the two different PLP variations.

No abstract available

This study employs the framework of multimodal cognitive adaptation theory to investigate personalized learning phenomena in AI-enhanced English classrooms at vocational undergraduate institutions. The theoretical analysis begins with a detailed exploration of the concept's core principles, followed by an examination of the theoretical foundations supporting AI-integrated teaching environments. Current practices reveal both distinctive features and limitations of personalized learning approaches. These findings establish crucial empirical groundwork for future research. Notably, the multimodal cognitive adaptation-oriented learning pathway design principles developed from these insights demonstrate significant theoretical value. Practical implementation strategies for constructing personalized learning pathways have been validated in vocational English education. The study provides dual benefits: it offers theoretical references for integrating AI into classroom teaching reforms, while also contributing to improving foreign language instruction quality and fostering learners' individualized language development in vocational colleges.

Badminton motor skills are crucial for hitting the ball. Colleges and universities should focus on using multimedia technology and AI feedback technology to assist badminton teaching, helping students master badminton skills and improve the effectiveness of badminton sports. Based on this, this paper studies the application of multimedia technology and AI feedback in the learning of college students’ badminton motor skills, expounds on its application value, and proposes specific application strategies. The aim is to help college students learn badminton motor skills efficiently and provide new ideas and practical references for reforming physical education teaching in colleges and universities.

In the conventional e-learning environment, most of the course recommendation system only utilizes academic information, which hinders the recommendations from developing with learners' skills and needs. This work presents SmartPath system, a lightweight and explainable Artificial Intelligence framework that generates personalized and dynamic learning paths. The SmartPath system assesses learner skill levels by using diagnostic quizzes, learner preferences, and learner previous performances and utilizing Decision Tree classifiers. The assessments lead to generating topic-based roadmaps utilizing Directed Acyclic graphs and recommending learning resources from open educational resources such as YouTube and W3Schools. The system uses GPT-based explainability to give a transparent rationale behind each recommendation and gamified elements such as badges and certificates to elevate learner motivation. The SmartPath System continues to gain learner feedback to dynamically adjust to recommendations. Findings from an experimental evaluation demonstrate that learners in Smart Path make significant improvements in knowledge and satisfaction than learners following conventional static paths. The findings reveal that SmartPath is a scalable, domain-independent framework for improving personalized e-learning experience.

No abstract available

With the rapid development of artificial intelligence (AI) technology, the teaching mode in the field of education is undergoing profound changes. Especially the design and implementation of personalized learning paths have become an important direction of intelligent teaching reform. The traditional “one-size-fits-all” teaching model has gradually failed to meet the individualized learning needs of students. However, through the advantages of data analysis and real-time feedback, AI technology can provide tailor-made teaching content and learning paths based on students’ learning progress, interests, and abilities. This study explores the innovation of the personalized learning path model based on AI technology, and analyzes the potential and challenges of this model in improving teaching effectiveness, promoting the all-round development of students, and optimizing the interaction between teachers and students. Through case analysis and empirical research, this paper summarizes the implementation methods of the AI-driven personalized learning path, the innovation of teaching models, and their application prospects in educational reform. Meanwhile, the research also discussed the ethical issues of AI technology in education, data privacy protection, and its impact on the teacher-student relationship, and proposed corresponding solutions.

This study explores the application of artificial intelligence (AI) and deep learning (DL) technologies in graduate education to promote the inheritance and development of the scientist spirit. This study employs a Long Short-Term Memory (LSTM) network to predict students' learning paths. Meanwhile, it constructs a DL-based personalized learning path and resource recommendation model by integrating a hybrid recommendation mechanism combining collaborative filtering and content-based filtering. The model inputs students' historical learning data and utilizes LSTM to capture long-term dependencies for predicting future learning activities. At the same time, it dynamically adjusts the learning rate through a reinforcement learning mechanism to optimize model performance. Additionally, this study introduces the Local Interpretable Model-Agnostic Explanations (LIME) algorithm to enhance the model's interpretability, ensuring that educators can understand the model's decision-making logic. Model training employs cross-validation techniques, and Principal Component Analysis (PCA) is used for dimensionality reduction and feature selection to improve data processing efficiency. Experimental results demonstrate that the DL model significantly outperforms traditional models in personalized learning path prediction, resource matching efficiency, and student performance prediction. Particularly, the DL model has an accuracy of 92.5%, an F1 score of 91.8%, an Area Under the Receiver Operating Characteristic Curve value of 0.95, a user satisfaction rate of 89.2%, and a prediction bias of only -0.75%. Furthermore, through user satisfaction surveys and expert reviews, this study qualitatively analyzes the impact of AI and DL technologies on educational practices. This confirms their value in enhancing education quality and fostering a scientist spirit. The study concludes that AI and DL technologies can effectively optimize graduate education models and promote the inheritance of the scientist spirit. Moreover, these technologies can cultivate innovative capabilities and provide theoretical support and practical guidance for intelligent educational reform.

Large Language Models (LLMs) hold immense potential for transforming education by automating the generation of personalized learning paths. However, traditional LLMs often suffer from hallucinations and content irrelevance. To address these challenges, we propose SKYRAG, a Separated Keyword Retrieval Augmentation Generation system that enhances the learning path generation process by integrating advanced retrieval mechanisms with LLMs. SKYRAG retrieves relevant course materials from Massive Open Online Course (MOOC) platforms, aligning them with individual learner profiles to provide personalized and coherent learning paths. Compared with Naïve RAG, SKYRAG demonstrates superior performance in terms of accuracy, relevance, and user satisfaction, as confirmed by human evaluations across four domains. By improving retrieval precision and addressing the limitations of traditional methods, SKYRAG represents a significant advancement in educational technology. This study contributes to the growing body of research on AI-driven learning systems and highlights SKYRAG’s potential for widespread adoption in dynamic educational environments.

With the development of large language model (LLM) technology, AI-assisted education systems are gradually being widely used. Learning Path Recommendation (LPR) is an important task in personalized instructional scenarios. AI-assisted LPR is gaining traction for its ability to generate learning content based on a student’s personalized needs. However, the native-LLM has the problem of hallucination, which may lead to the inability to generate learning content; in addition, the evaluation results of the LLM on students’ knowledge status are usually conservative and have a large margin of error. To address these issues, this work proposes a novel approach for LPR enhanced by knowledge tracing (KT) and LLM. Our method operates in a “generate-and-retrieve” manner: the LLM acts as a pedagogical planner that generates contextual reference exercises based on the student’s needs. Subsequently, a retrieval mechanism constructs the concrete learning path by retrieving the top-N most semantically similar exercises from an established exercise bank, ensuring the recommendations are both pedagogically sound and practically available. The KT plays the role of an evaluator in the iterative process. Rather than generating semantic instructions directly, it provides a quantitative and structured performance metric. Specifically, given a candidate learning path generated by the LLM, the KT model simulates the student’s knowledge state after completing the path and computes a knowledge promotion score. This score quantitatively measures the effectiveness of the proposed path for the current student, thereby guiding the refinement of subsequent recommendations. This iterative interaction between the KT and the LLM continuously refines the candidate learning items until an optimal learning path is generated. Experimental validations on public datasets demonstrate that our model surpasses baseline methods.

OMO teaching mode based on artificial intelligence big model is one of the important future research directions and application landing forms in the future education field. The learning path recommendation algorithm based on big language model is constructed by integrating Transformer architecture, neural network architecture and self-attention mechanism. Combining it with the course knowledge graph, it links the learners with the knowledge system and visualizes the results of the intelligently planned learning path. The study shows that compared with several other algorithms, the personalized learning path recommendation algorithm based on AI big model has better convergence speed and stability. The optimal solution for learning path planning is found after only about 90 iterations. Taking “Chemical Process and Control Simulation” as the target course, the method in this paper gives the learning path and course. Through the questionnaire survey, the mean value of the four dimensions of pre-class pre-study, classroom exploration, post-class enhancement, and learning satisfaction is more than 3 points, which indicates that the OMO model and the teaching model of the artificial intelligence big model have a better experience.

: In this study, 100 new employees were randomly divided into experimental group (AI-based personalized learning path) and control group (traditional skills training path) for a three-month comparative experiment. The experimental group collected learner data through AI technology, customized personalized learning plans, recommended learning resources, and provided immediate feedback and evaluation. The control group received traditional training with unified curriculum and fixed teaching resources. The results show that the experimental group is significantly better than the control group in comprehensive skills test scores, task completion rate, learning satisfaction, learning motivation and participation. Statistical analysis shows that there are significant differences in learning progress, achievements and attitudes between the experimental group and the control group, and the personalized learning path has a significant effect on improving the learning effect of employees. Correlation analysis further reveals the positive correlation between learning time, skill test scores and employee satisfaction and participation. This study confirms the potential of AI-based personalized learning path in improving the effect of skills training, and provides a useful reference for the further development and application of personalized learning in the field of education in the future.

This paper discusses the application of intelligent algorithm in students' personalized learning path planning, aiming at improving learning efficiency and optimizing the allocation of teaching resources through AI technology. Firstly, this paper introduces the importance of personalized learning and points out the limitations of traditional teaching mode. Then, it discusses in detail how the intelligent algorithm collects and analyzes students' learning data, including grades, behavior tracks, interest preferences and emotional States, to construct students' learning portraits. On this basis, this paper proposes an intelligent algorithm framework based on reinforcement learning (RL), which uses K-means clustering algorithm, decision tree model and markov decision processes (MDP) to design and optimize personalized learning paths. Through the Q-learning algorithm, the agent can choose the best learning resources or activities according to the students' learning state, and generate a coherent learning path that conforms to the students' preferences. The case study shows that the framework has significantly improved students' learning efficiency and grades, reduced learning time, and improved students' satisfaction and participation in senior high school mathematics courses.

In recent years, advances in technology and the increased use of online and distance learning have led to the widespread adoption of Learning Management Systems (LMS) in higher education institutions and in software engineering education. However, today’s LMS typically consider only a few learner characteristics when recommending personalized learning paths. It is evident that relying on a single learning theory or a limited set of learner characteristics does not fully capture the complex behavior and needs of learners. To address this, we present a hybrid AI algorithm called Nestor that integrates qualitative insights and quantitative evidence to incorporate multiple learning theories, including learning styles, learning strategies, personalities, and preferences for learning elements. In this paper we discuss the design of Nestor to defining its structure and conduct an in-depth comparative analysis to identify optimal data for parameter learning. In addition, a qualitative questionnaire survey of learners who received learning paths evaluates the effectiveness of the recommendations. Performance evaluations on held-out test data and leave-one-out cross-validation indicate that Nestor explains both empirical and synthesized data similarly while achieving improved predictive performance with augmented datasets. Moreover, learner feedback shows neutral to positive responses with the recommendations.

Massive Open Online Courses (MOOCs) generate vast amounts of learner interaction data, enabling predictive models to identify at-risk students. However, most existing approaches stop at prediction, offering limited guidance on which interventions should be applied to improve outcomes. This paper proposes a Causal-AI–driven reinforcement learning (RL) framework that transitions from prediction to prescription by integrating causal effect estimation with policy optimization. Specifically, uplift modeling is employed to estimate heterogeneous treatment effects of candidate learning actions, and these causal signals are embedded into an Actor–Critic RL agent through causal regularization. The resulting framework prescribes adaptive intervention policies that are both data-driven and causally grounded. We evaluate the approach on the OULAD dataset. Baseline predictive models achieve strong accuracy (AUC = 0.795) and calibration, while uplift distributions reveal heavy-tailed heterogeneity across learners. Off-policy evaluation demonstrates that the causal-RL policy consistently outperforms the observed behavior and standard RL baselines (Greedy Uplift SNIPS $\approx 0.619$ vs. baseline 0.589). Subgroup analysis highlights disparities across age and gender, motivating fairness-aware extensions. A case study further illustrates how adaptive recommendations can redirect learner trajectories toward success. These findings demonstrate that causal-RL integration enables prescriptive, interpretable, and equitable personalization in MOOCs.

: This study takes the course of Macroeconomics as a practical carrier, focusing on the deep integration of AI technology and the "teaching learning evaluation" system of higher education, and exploring the value-added logic and implementation path of students' professional competence. By conducting literature research to clarify core concepts and theoretical foundations, and combining empirical research methods to compare and analyze the differences in effectiveness between AI driven "teaching learning evaluation" closed-loop and traditional teaching models, a theoretical model of "AI empowering teaching learning evaluation → closed - loop dynamic linkage → professional competence enhancement" is constructed, and four major paths for professional competence enhancement are designed. The research results indicate that there has been a significant improvement in the four dimensions of professional cognition, professional practice, innovative thinking, and independent reflection. Research can provide a practical paradigm for the cultivation of professional competence in economic management courses, enriching the theoretical research on the integration of AI education and professional competence.

This study proposes a multi-agent collaboration system based on the low-code Coze platform to enhance students’ English third classroom practices. Addressing the limitations of traditional extracurricular learning—including fragmented scenarios, insufficient personalized guidance, and a lack of virtual-real coordination—we leverage Coze’s visual workflow engine to integrate training in listening, speaking, reading, and writing. By incorporating technologies such as speech recognition, multimodal generation, and dynamic knowledge graphs, the system provides adaptive learning support. Case studies across four language skill scenarios demonstrate its effectiveness in improving training coherence, feedback immediacy, and cross-scenario adaptability. This potential platform boosts teacher-student engagement, offering a replicable technical solution for educational digital transformation.

Education is a cornerstone of societal progress, continually evolving to meet the needs of diverse learners. In our daily lives, technology has become ubiquitous, enhancing communication, efficiency, and access to information. This technological integration extends to education, where digital tools are increasingly used to support and enhance learning experiences. In the educational landscape, technology plays a vital role by providing interactive learning platforms, digital resources, and tools that cater to various learning styles. Artificial Intelligence further revolutionizes this field by enabling personalized learning experiences. AI-driven applications analyze student data to tailor educational content, identify strengths and weaknesses, and offer targeted support, thereby promoting engagement and academic success. This work focuses on researching, investigating, and implementing a new learning experience based on the analysis of the learning outcomes. It highlights the importance of the role of Artificial Intelligence (AI) in enhancing the learning process for university students. It tries to integrate AI in the learning process to personalize the students’ learning path by focusing on the progress shown on a customized analytics dashboard. We aim to create a tool that empowers students to take ownership of their educational journey.

In response to the pain points of rapid iteration of front-end education technology, large differences in learner foundations, and a lack of practical scenarios, this paper combines generative artificial intelligence and AI agents to analyze the empowerment logic from three dimensions: knowledge ecology reconstruction, cognitive collaborative upgrading, and teaching methodology innovation. It explores its application scenarios in teaching and learning, sorts out challenges such as technology adaptation and learning dependence, and proposes paths such as building an exclusive AI ecosystem and optimizing the guidance mechanism of intelligent agents to provide support for the digital transformation of front-end education.

As a fundamental expression of traditional Chinese culture, Sinology has become an integral component of Chinese language and cultural education programs for international students. However, two key challenges constrain effective Sinology learning: the abstract nature of classical Chinese cultural concepts and the gap between modern pedagogical methods and historical contexts. To address these issues, this paper presents an immersive Sinology learning framework that integrates Extended Reality (XR) technology with AI Agents. The framework aims to bridge cultural and temporal divides in contemporary Sinology instruction. Using the cultural interpretation of the Northern Song Dynasty masterpiece Along the River During the Qingming Festival (adapted for international learners) as a case study, the framework constructs a virtual environment that accurately recreates the urban landscape, social interactions, and cultural features of the Northern Song period. Through XR devices, international students can virtually experience this historical setting as though traveling through time and space. Within the environment, AI-driven virtual characters support real-time multilingual communication enabled by advanced speech recognition technology. These interactions allow students to acquire a deeper understanding of Song Dynasty culture while simultaneously learning Chinese expressions related to the specific scenes, thereby realizing the integrated learning of cultural cognition and language use. To assess the framework's effectiveness, a controlled experiment was carried out with 20 international students possessing intermediate Chinese proficiency. The experimental group utilized the XR-AI Sinology learning system, whereas the control group participated in conventional text-based lectures supplemented by classroom instruction on content derived from Along the River During the Qingming Festival.Learning outcomes were assessed using pre-test and post-test measurements across three key dimensions—cultural knowledge, mastery of Chinese phrases, and learning satisfaction. The collected data were statistically analyzed using t-tests. The results revealed that the experimental group achieved a significantly greater improvement in cultural knowledge scores, with a 131.6% increase in the post-test (p < 0.001), compared to a 51.3% gain in the control group (p = 0.012). In terms of Chinese phrase mastery, the experimental group also outperformed the control group, attaining a mean post-test score of 56.50 ± 11.87, as opposed to 38.90 ± 10.56 in the control group. Furthermore, in the learning satisfaction survey, the experimental group reported higher ratings across all evaluated aspects—including content attractiveness and interaction experience—than their counterparts in the control group.Overall, the experimental group demonstrated statistically significant superiority over the control group in all three evaluated metrics (p < 0.05). These findings substantiate that the XR-AI Sinology framework significantly enhances both international students’ engagement and learning outcomes in Sinology education.

In today's rapidly evolving education field, generative artificial intelligence (AI) technology is increasingly becoming a key force in promoting teaching innovation and personalized learning solutions. This article deeply explores the application of generative AI technology in designing customized learning paths in higher education and its profound impact on educational practice. Through specific case analysis, this study reveals how generative AI can effectively analyze students' learning behaviors, preferences, and performance to automatically generate and adjust learning paths in real time, thereby providing students with a highly personalized learning experience. Research results show that using generative AI technology can significantly improve learning efficiency, increase student engagement and satisfaction, and optimize learning outcomes. In addition, this article also demonstrates the ability of generative AI to promote innovation in learning content and teaching methods, improve the efficiency of educational resource utilization, and enhance the adaptability of learning paths. Through these findings, this paper provides valuable insights for higher education institutions, pointing out the importance and feasibility of leveraging generative AI technologies to design personalized learning paths, while highlighting key considerations that should be noted during implementation.

Abstract The advent of informatization and intelligent systems in education holds substantial practical importance. This study focuses on evaluating the application effects of an AI tutoring model in extralinguistic learning, using a middle school as a case study. Two parallel classes were selected to design teaching experiments aimed at assessing the AI tutoring model’s efficacy by comparing students’ language performance before and after its implementation. Furthermore, factors influencing the willingness to use the AI tutoring model were identified, leading to the construction of a structural equation model and the formulation and testing of related hypotheses. Results indicate that the AI tutoring model significantly enhances students’ language learning outcomes, with a performance increase of 12.02 points post-instruction, which is 10.19 points higher than that observed in the control group. Variables such as performance expectation, hedonic motivation, willingness to communicate, and interaction with AI demonstrated a positive and significant impact on attitudes (p < 0.05). They indirectly fostered a sustained desire to engage with the AI tutoring model. This research offers valuable insights for the future effective integration of AI into extralinguistic educational contexts.

In the era of rapidly evolving technological advancements and the growing importance of 21st-century skills and environmental consciousness, education systems face the challenge of providing personalized learning experiences that cater to individual student needs while fostering critical thinking, problem-solving abilities, and environmental awareness. This paper presents a formal model for personalizing learning paths through the integration of artificial intelligence (AI) into in- structional planning. The proposed model accounts for varying difficulty levels in learning activities related to 21st-century skills and environmental care, creating a comprehensive framework for optimizing student learning paths.

MOOCs has a great impact on nowadays educational strategies. MOOCs enable global learners to learn without time and space constraints, allowing distinct learning characteristics when participating in online courses. Overwhelmed by complicated learning resources, a problem named “information overload” was widely discussed in online education. AI-based Recommender System, which is recognized as the powerful solution to improve resource acquisition via customized supply, has been regarded as an assistant in online learning by giving personalized learning strategies. In this paper, a Personalized Learning Path Recommender System with LINE Bot is proposed to meet personal preferences on path of learning. A LSTM model is built to consider video-watching preference features, clusters of students and learning paths to recommend personal learning path suitable for each student. Related recommendation contents and prediction results will be received by users through in-time LINE massages, achieving the goal of making in-time and active recommendations. From the evaluation part, F1-score of the proposed Learning Path Prediction Model is 0.8, indicating this model has a certain degree of accuracy. On the other hand, the proposed system is used in two courses of NTHU Cloud to give personalized learning path guidance. The experimental results demonstrate that learning path recommendations will help students have stronger learning willingness to keep learning, and help plan proper study steps to fulfill their own learning needs. On the other hand, this system provides another way except examinations to make judgements about one's learning status, and most learners agree that this kind of recommendations is helpful to review unfamiliar concepts and catch up with others.

No abstract available

Learning Path Recommendation is the heart of adaptive learning, the educational paradigm of an Interactive Educational System (IES) providing a personalized learning experience based on the student's history of learning activities. In typical existing IESs, the student must fully consume a recommended learning item to be provided a new recommendation. This workflow comes with several limitations. For example, there is no opportunity for the student to give feedback on the choice of learning items made by the IES. Furthermore, the mechanism by which the choice is made is opaque to the student, limiting the student's ability to track their learning. To this end, we introduce Rocket, a Tinder-like User Interface for a general class of IESs. Rocket provides a visual representation of Artificial Intelligence (AI)-extracted features of learning materials, allowing the student to quickly decide whether the material meets their needs. The student can choose between engaging with the material and receiving a new recommendation by swiping or tapping. Rocket offers the following potential improvements for IES User Interfaces: First, Rocket enhances the explainability of IES recommendations by showing students a visual summary of the meaningful AI-extracted features used in the decision-making process. Second, Rocket enables self-personalization of the learning experience by leveraging the students' knowledge of their own abilities and needs. Finally, Rocket provides students with fine-grained information on their learning path, giving them an avenue to assess their own skills and track their learning progress. We present the source code of Rocket, in which we emphasize the independence and extensibility of each component, and make it publicly available for all purposes.

This paper establishes a specific path for the realization of AI-enhanced learning on the content of Civic and Political Education, starting from the relevance, quality, novelty and intuitiveness of the teaching content. Through HTML parsing and other crawler technology to obtain the Civics education data on the news network, and extract the data characteristics of the Civics material, using the clustering rule algorithm, to classify the material. Decision tree calculation based on random forest is performed to dynamically expand and integrate the material, on this basis, using reinforcement learning recommendation algorithm, the Civic and political education content recommendation model is constructed, and the recommendation results of the algorithm are verified using simulation experiments. The experimental results show that the average success rate of the research-designed recommendation algorithm in the last 10 groups of experimental data is 25.218%, which is higher than that of the MK recommendation algorithm (18.03%), and the average time of the research-designed recommendation algorithm in the last 10 groups of data is 5.095s, which is more efficient than that of the MK recommendation algorithm (11.903s). After integrating the enhanced learning content recommendation in the Civics education, the students' humanism scale score was 100.56±12.364, with a p-value of less than 0.05, which was significantly higher than that before teaching.