Explainable AI in Education

Journal Name	Impact Factor	CiteScore	Launched Year	First Decision (median)	APC
AI ai	6.5	7.3	2020	20.4 Days	CHF 1800
Applied Sciences applsci	2.9	6.1	2011	15 Days	CHF 2400
Education Sciences education	3.5	6.2	2011	24.8 Days	CHF 2000
Electronics electronics	2.9	7.0	2012	14.8 Days	CHF 2400
Information information	4.3	8.2	2010	18.7 Days	CHF 1800

30 pages, 1001 KB

Open AccessArticle

CADKT: A Cognitive-Affective Dual-Graph Contrastive Framework for Dynamic Affect-Aware Knowledge Tracing

by Hongwei Wang, Liqing Qiu and Mingxiang He

Electronics 2026, 15(15), 3433; https://doi.org/10.3390/electronics15153433 - 3 Aug 2026

Viewed by 366

Knowledge tracing (KT) predicts students’ future performance by modeling their knowledge mastery from historical learning interactions. Although recent deep learning-based KT models have achieved promising results, most of them mainly rely on response sequences or cognitive structures, leaving affect-related learning states insufficiently explored. [...] Read more.

Knowledge tracing (KT) predicts students’ future performance by modeling their knowledge mastery from historical learning interactions. Although recent deep learning-based KT models have achieved promising results, most of them mainly rely on response sequences or cognitive structures, leaving affect-related learning states insufficiently explored. In real learning scenarios, student performance is shaped not only by cognitive mastery but also by affect-related factors that change during the learning process. To address this issue, this paper proposes a Cognitive-Affective Dual-Graph Knowledge Tracing framework, termed CADKT, which jointly models cognitive structure and behavior-derived affect-related dynamics. Specifically, CADKT first extracts affect-related behavioral factors from learning logs, including correctness discrepancy, time efficiency discrepancy, and learning persistence deviation. These factors are then used to infer interval-level behavior-derived affect-related profiles without relying on explicit emotion annotations, and the inferred profiles are organized by a Temporal Affective Graph to model their transitions along the learning sequence. Meanwhile, a Heterogeneous Knowledge Graph is constructed to represent student–problem interactions and problem–concept associations. To integrate the two views, CADKT introduces cross-view contrastive regularization and a Projected Hybrid Softmax fusion mechanism for next-response prediction. Experiments on four public educational datasets show that CADKT achieves stronger overall performance than representative KT baselines on most evaluation metrics. Further ablation studies, profile analysis, and case visualization demonstrate that CADKT improves prediction performance while providing interpretable evidence for the interaction between knowledge mastery and behavior-derived affect-related dynamics. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

22 pages, 4150 KB

Open AccessArticle

Hybrid Career Path Recommendation System for Computer Science Students: Integrating Academic Profiles and Personal Skills with a Content–Collaborative Model

by Taif Almufareh and Abdulaziz Almaleh

Electronics 2026, 15(14), 3228; https://doi.org/10.3390/electronics15143228 - 22 Jul 2026

Viewed by 582

Abstract

Career decision-making remains a critical challenge for computer science students, particularly within rapidly evolving labor markets such as Saudi Arabia. Traditional guidance approaches often rely on subjective advice or limited academic indicators, which frequently result in mismatched career choices and reduced employability. To [...] Read more.

Career decision-making remains a critical challenge for computer science students, particularly within rapidly evolving labor markets such as Saudi Arabia. Traditional guidance approaches often rely on subjective advice or limited academic indicators, which frequently result in mismatched career choices and reduced employability. To address this gap, we propose a hybrid career recommendation system that integrates academic performance, technical and soft skills, and personal preferences. The proposed system combines artificial neural network (ANN) prediction, content-based career-profile matching, and collaborative peer similarity through weighted score fusion to generate ranked career path recommendations. The content-based component employs a similarity function (cosine similarity) to measure the alignment between the student’s attributes and each career profile. The collaborative component applies a k-Nearest Neighbors model to capture interaction-like signals from peer profiles. The ANN learns complex nonlinear relationships between student features and career outcomes, enabling the system to generate predictive probabilities for each career path by applying the TensorFlow framework for multi-class classification of student data. Final recommendations are produced through a weighted fusion of models. An experimental evaluation of 400 student records using stratified macro-level evaluation metrics demonstrates robust performance, achieving an accuracy of 93.75%, precision of 0.94, recall of 0.94, and F1-score of 0.94. The system provides Top-k-ranked career suggestions, along with interpretable feature importance, enhancing transparency for academic advisors and students. The results highlight the effectiveness of hybrid recommenders in mitigating cold-start and sparsity problems while capturing multidimensional student attributes. This study contributes a reproducible methodology and implementation pipeline that can be extended to real institutional contexts. Future work will include expansion to more university disciplines and integration with academic consulting platforms to support data-driven career planning. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

23 pages, 4720 KB

Open AccessArticle

Explainable Artificial Intelligence (XAI) for Identifying the Integration of International Students in the Host Country and Its Culture

by James Vakilian, Fareed Ud Din, Edmund J. Sadgrove, Mohammadreza Haghighat and Niusha Shafiabady

AI 2026, 7(7), 238; https://doi.org/10.3390/ai7070238 - 25 Jun 2026

Cited by 1 | Viewed by 683

Abstract

The integration of international students into host countries and their cultures is a multifaceted challenge that significantly impacts their academic success and well-being. This study leverages Explainable Artificial Intelligence (XAI) to model and interpret variables associated with the self-rated integration of 175 international [...] Read more.

The integration of international students into host countries and their cultures is a multifaceted challenge that significantly impacts their academic success and well-being. This study leverages Explainable Artificial Intelligence (XAI) to model and interpret variables associated with the self-rated integration of 175 international students at Charles Darwin University (CDU) in Australia, using data from a 42-question survey. Employing machine learning models, including Decision Tree (DT) and Gradient Boosting Machine (GBM), we use XAI techniques to identify variables most strongly associated with students’ self-rated integration, including career confidence, perceived future happiness, and perceived career obstacles. SHAP analyses and partial dependence plots provide global and instance-level insights, revealing both the magnitude and directional effects of these features. The findings highlight the predictive relevance of psychological and social variables in students’ self-rated integration, offering exploratory insights that inform targeted support programs. By enhancing model transparency through XAI, this research fosters trust in AI-driven educational interventions, addressing ethical considerations and promoting equitable outcomes for diverse student populations. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

23 pages, 1163 KB

Open AccessArticle

Personalized Course Recommendation Based on Attribute-Interaction Joint Encoding and Hypergraph Reconstruction

by Jun Yi, Xiaoqi Han, Wei Zhou, Shan Xiao and Ming Liu

Information 2026, 17(6), 598; https://doi.org/10.3390/info17060598 - 15 Jun 2026

Viewed by 364

Abstract

Course recommendation systems based on deep learning have demonstrated powerful feature extraction capabilities in dealing with information overload in massive open online courses (MOOCs), and have become an irreplaceable mainstream method. However, the learner–course interactions are usually scarce in reality, which limits the [...] Read more.

Course recommendation systems based on deep learning have demonstrated powerful feature extraction capabilities in dealing with information overload in massive open online courses (MOOCs), and have become an irreplaceable mainstream method. However, the learner–course interactions are usually scarce in reality, which limits the representation power of course recommendation. In addition, the contribution of learner and course attribute information to course recommendation has not been sufficiently explored by most existing methods. To tackle these challenges, a personalized course recommendation model based on attribute-interaction joint encoding and hypergraph reconstruction (AIHR-PCRM) is proposed in this paper. Specifically, a course hypergraph reconstruction (CHR) method is designed to construct higher-order associations for each course to explore more reliable global collaboration signals. Unlike existing hypergraph constructions that directly take learners as hyperedges, CHR explicitly couples three steps, including invalid learner elimination, high-order reachability induction, and similarity-based hyperedge filtering, to substantially raise the signal-to-noise ratio of the resulting hypergraph. Based on this, a hypergraph global collaborative learning module (HGM) can alleviate the issue of data sparsity. Then, a joint encoding module (JEM) is utilized to enhance learner behavior sequence representations by simultaneously fusing hypergraph-level global signals with attribute-level local semantics. Finally, a bidirectional self-attention module (BSM) is introduced to blend the contextual information of the learner behavior sequence, and to further provide a recommendation. Experimental results on three real-world datasets revealed that the proposed model has already achieved the best recall and ndcg scores compared to those of several existing models. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

28 pages, 3558 KB

Open AccessSystematic Review

AI-Based Academic Advising Across the Student Lifecycle: A Systematic Literature Review

by Ilyas Alloug, Mohamed Daoudi and Ilham Oumaira

Information 2026, 17(4), 335; https://doi.org/10.3390/info17040335 - 1 Apr 2026

Viewed by 2097

Abstract

Academic advising is fundamental to student success, yet the rapid integration of Artificial Intelligence (AI) and Machine Learning (ML) is fundamentally transforming the delivery of academic support. While predictive models and Recommendation Systems (RS) are becoming more accessible, the existing literature remains fragmented [...] Read more.

Academic advising is fundamental to student success, yet the rapid integration of Artificial Intelligence (AI) and Machine Learning (ML) is fundamentally transforming the delivery of academic support. While predictive models and Recommendation Systems (RS) are becoming more accessible, the existing literature remains fragmented across diverse technical architectures and institutional objectives, preventing a clear understanding of the field’s evolution. In view of this, we present a Systematic Literature Review (SLR) of AI-driven academic advising, adhering to the PRISMA 2020 framework. We analyzed 27 peer-reviewed studies published between 2018 and 2025 to synthesize methodological trends and functional applications. Our findings reveal that while most systems prioritize pathway recommendations via classical ML or hybrid architectures, Early-Warning Systems (EWS) remain anchored in predictive classification. Furthermore, a nascent shift toward Generative AI (GenAI) indicates a move toward more interactive advising, though transparency and evaluation standards remain inconsistent. This review identifies a critical tension between algorithmic performance and institutional interpretability. We conclude by proposing a research agenda that emphasizes the need for cross-context validation and the development of socio-technical frameworks that integrate AI into existing higher education management structures. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

22 pages, 3647 KB

Open AccessArticle

Addressing Class Imbalance in Predicting Student Performance Using SMOTE and GAN Techniques

by Fatema Mohammad Alnassar, Tim Blackwell, Elaheh Homayounvala and Matthew Yee-king

Appl. Sci. 2026, 16(7), 3274; https://doi.org/10.3390/app16073274 - 28 Mar 2026

Viewed by 1094

Abstract

Virtual Learning Environments (VLEs) have become increasingly popular in education, particularly with the rise of remote learning during the COVID-19 pandemic. Assessing student performance in VLEs is challenging, and the accurate prediction of final results is of great interest to educational institutions. Machine [...] Read more.

Virtual Learning Environments (VLEs) have become increasingly popular in education, particularly with the rise of remote learning during the COVID-19 pandemic. Assessing student performance in VLEs is challenging, and the accurate prediction of final results is of great interest to educational institutions. Machine learning classification models have been shown to be effective in predicting student performance, but the accuracy of these models depends on the dataset’s size, diversity, quality, and feature type. Class imbalance is a common issue in educational datasets, but there is a lack of research on addressing this problem in predicting student performance. In this paper, we present an experimental design that addresses class imbalance in predicting student performance by using the Synthetic Minority Over-sampling Technique (SMOTE) and Generative Adversarial Network (GAN) technique. We compared the classification performance of seven machine learning models (i.e., Multi-Layer Perceptron (MLP), Decision Trees (DT), Random Forests (RF), Extreme Gradient Boosting (XGBoost), Categorical Boosting (CATBoost), K-Nearest Neighbors (KNN), and Support Vector Classifier (SVC)) using different dataset combinations, and our results show that SMOTE techniques can improve model performance, and GAN models can generate useful simulated data for classification tasks. Among the SMOTE resampling methods, SMOTE NN produced the strongest performance for the RF model, achieving a Region of Convergence (ROC) Area Under the Curve (AUC) of 0.96 and a Type II error rate of 8%. For the generative data experiments, the XGBoost model demonstrated the best performance when trained on the GAN-generated dataset balanced using SMOTE NN, attaining a ROC AUC of 0.97 and a reduced Type II error rate of 3%. These results indicate that the combined use of class balancing techniques and generative synthetic data augmentation can enhance student outcome prediction performance. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

15 pages, 1495 KB

Open AccessPerspective

Artificial Intelligence in Higher Education: A Global Statistical Synthesis for Policy and Quality Assurance Reform

by Rima J. Isaifan

Educ. Sci. 2026, 16(3), 483; https://doi.org/10.3390/educsci16030483 - 20 Mar 2026

Cited by 3 | Viewed by 3464

Abstract

Artificial intelligence has transitioned from a peripheral innovation to a core infrastructure shaping higher education within a remarkably short period. While conceptual debates on AI ethics, pedagogy, and academic integrity are expanding, empirically grounded syntheses that consolidate global evidence remain limited. This study [...] Read more.

Artificial intelligence has transitioned from a peripheral innovation to a core infrastructure shaping higher education within a remarkably short period. While conceptual debates on AI ethics, pedagogy, and academic integrity are expanding, empirically grounded syntheses that consolidate global evidence remain limited. This study addresses this gap by providing an integrated cross-domain synthesis and statistically grounded overview of AI adoption, use, and governance across higher education systems. Using a secondary statistical synthesis methodology, the study aggregates large-scale quantitative data published between 2021 and 2025 from reputable international sources, including student and faculty surveys, institutional reports, research indices, and regulatory datasets. Results demonstrate near-universal student adoption of AI tools, rapid but uneven professional engagement among faculty and staff, a sharp rise in AI-related academic misconduct, accelerating impacts on research production and scientific workflows, and persistent gaps in institutional preparedness, policy development, and equity. The findings reveal a widening disconnect between bottom-up AI adoption and top-down governance mechanisms, particularly in assessment design, academic integrity frameworks, faculty capacity building, and quality assurance systems. Moreover, this paper argues that AI can no longer be treated as an optional educational technology and must instead be governed as a foundational component of higher education infrastructure. The study concludes by outlining evidence-based policy implications for institutions, regulators, and quality assurance agencies, emphasizing the need for coordinated, adaptive, and equity-oriented governance frameworks grounded in empirical realities rather than speculative narratives. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

29 pages, 1545 KB

Open AccessArticle

Hierarchical Aggregation of Local Explanations for Student Adaptability

by Leonard Chukwualuka Nnadi and Yutaka Watanobe

Appl. Sci. 2026, 16(1), 333; https://doi.org/10.3390/app16010333 - 29 Dec 2025

Cited by 4 | Viewed by 1042

Abstract

In this study, we present Hierarchical Local Interpretable Model-agnostic Explanations (H-LIME), an innovative extension of the LIME technique that provides interpretable machine learning insights across multiple levels of data hierarchy. While traditional local explanation methods focus on instance-level attributions, they often overlook systemic [...] Read more.

In this study, we present Hierarchical Local Interpretable Model-agnostic Explanations (H-LIME), an innovative extension of the LIME technique that provides interpretable machine learning insights across multiple levels of data hierarchy. While traditional local explanation methods focus on instance-level attributions, they often overlook systemic patterns embedded within educational structures. To address this limitation, H-LIME aggregates local explanations across hierarchical layers, Institution Type, Location, and Educational Level, thereby linking individual predictions to broader, policy-relevant trends. We evaluate H-LIME on a student adaptability dataset using a Random Forest model chosen for its superior explanation stability (approximately 4.5 times more stable than Decision Trees). The framework uncovers consistent global predictors of adaptability, such as education level and class duration, while revealing subgroup-specific factors, including network type and financial condition, whose influence varies across hierarchical contexts. This work demonstrates the effectiveness of H-LIME at uncovering multi-level patterns in educational data and its potential for supporting targeted interventions, strategic planning, and evidence-based decision-making. Beyond education, the hierarchical approach offers a scalable solution for enhancing interpretability in domains where structured data relationships are essential. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

27 pages, 357 KB

Open AccessArticle

Ethical and Responsible AI in Education: Situated Ethics for Democratic Learning

by Sandra Hummel

Educ. Sci. 2025, 15(12), 1594; https://doi.org/10.3390/educsci15121594 - 26 Nov 2025

Cited by 11 | Viewed by 3334

Abstract

As AI systems increasingly structure educational processes, they shape not only what is learned, but also how epistemic authority is distributed and whose knowledge is recognized. This article explores the normative and technopolitical implications of this development by examining two prominent paradigms in [...] Read more.

As AI systems increasingly structure educational processes, they shape not only what is learned, but also how epistemic authority is distributed and whose knowledge is recognized. This article explores the normative and technopolitical implications of this development by examining two prominent paradigms in AI ethics: Ethical AI and Responsible AI. Although often treated as synonymous, these frameworks reflect distinct tensions between formal universalism and contextual responsiveness, between rule-based evaluation and governance-oriented design. Drawing on deontology, utilitarianism, responsibility ethics, contract theory, and the capability approach, the article analyzes the frictions that emerge when these frameworks are applied to algorithmically mediated education. The argument situates these tensions within broader philosophical debates on technological mediation, normative infrastructures, and the ethics of sociotechnical design. Through empirical examples such as algorithmic grading and AI-mediated admissions, the article shows how predictive systems embed values into optimization routines, thereby reshaping educational space and interpretive agency. In response, it develops the concept of situated ethics, emphasizing epistemic justice, learner autonomy, and democratic judgment as central criteria for evaluating educational AI. To clarify what is at stake, the article distinguishes adaptive learning optimization from education as a process of subject formation and democratic teaching objectives. Rather than viewing AI as an external tool, the article conceptualizes it as a co-constitutive actor within pedagogical practice. Ethical reflection must therefore be integrated into design, implementation, and institutional contexts from the outset. Accordingly, the article offers (1) a conceptual map of ethical paradigms, (2) a criteria-based evaluative lens, and (3) a practice-oriented diagnostic framework to guide situated ethics in educational AI. The paper ultimately argues for an approach that attends to the relational, political, and epistemic dimensions of AI systems in education. Full article

(This article belongs to the Topic Explainable AI in Education)

21 pages, 1133 KB

Open AccessArticle

Research on China’s Innovative Cybersecurity Education System Oriented Toward Engineering Education Accreditation

by Yimei Yang, Jinping Liu and Yujun Yang

Information 2025, 16(8), 645; https://doi.org/10.3390/info16080645 - 29 Jul 2025

Cited by 1 | Viewed by 3364

Abstract

This study, based on engineering education accreditation standards, addresses the supply–demand imbalance in China’s cybersecurity talent cultivation by constructing a sustainable “education-industry-society” collaborative model. Through case studies at Huaihua University and other institutions, employing methods such as literature analysis, field research, and empirical [...] Read more.

This study, based on engineering education accreditation standards, addresses the supply–demand imbalance in China’s cybersecurity talent cultivation by constructing a sustainable “education-industry-society” collaborative model. Through case studies at Huaihua University and other institutions, employing methods such as literature analysis, field research, and empirical investigation, we systematically explore reform pathways for an innovative cybersecurity talent development system. The research proposes a “three-platform, four-module” practical teaching framework, where the coordinated operation of the basic skills training platform, comprehensive ability development platform, and innovation enhancement platform significantly improves students’ engineering competencies (practical courses account for 41.6% of the curriculum). Findings demonstrate that eight industry-academia practice bases established through deep collaboration effectively align teaching content with industry needs, substantially enhancing students’ innovative and practical abilities (172 national awards, 649 provincial awards). Additionally, the multi-dimensional evaluation mechanism developed in this study enables a comprehensive assessment of students’ professional skills, practical capabilities, and innovative thinking. These reforms have increased the employment rate of cybersecurity graduates to over 90%, providing a replicable solution to China’s talent shortage. The research outcomes offer valuable insights for discipline development under engineering education accreditation and contribute to implementing sustainable development concepts in higher education. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

17 pages, 430 KB

Open AccessArticle

Adoption and Impact of ChatGPT in Computer Science Education: A Case Study on a Database Administration Course

by Daniel López-Fernández and Ricardo Vergaz

AI 2024, 5(4), 2321-2337; https://doi.org/10.3390/ai5040114 - 11 Nov 2024

Cited by 7 | Viewed by 4821

Abstract

The irruption of GenAI such as ChatGPT has changed the educational landscape. Therefore, methodological guidelines and more empirical experiences are needed to better understand these tools and know how to use them to their fullest potential. This contribution presents an exploratory and correlational [...] Read more.

The irruption of GenAI such as ChatGPT has changed the educational landscape. Therefore, methodological guidelines and more empirical experiences are needed to better understand these tools and know how to use them to their fullest potential. This contribution presents an exploratory and correlational study conducted with 37 computer science students who used ChatGPT as a support tool to learn database administration. The article addresses three questions: The first one explores the degree of use of ChatGPT among computer science students to learn database administration, the second one explores the profile of students who get the most out of tools like ChatGPT to deal with database administration activities, and the third one explores how the utilization of ChatGPT can impact in academic performance. To empirically shed light on these questions the student’s grades and a comprehensive questionnaire were employed as research instruments. The obtained results indicate that traditional learning resources, such as teacher’s explanations and student’s reports, were widely used and correlated positively with student’s grades. The usage and perceived utility of ChatGPT were moderate, but positive correlations between students’ grades and ChatGPT usage were found. Indeed, a significantly higher use of this tool was identified among the group of outstanding students. This indicate that high-performing students are the ones who are using ChatGPT the most. So, a new digital trench could be rising between these students and those with a lower degree of fundamentals and worse prompting skills, who may not take advantage of all the ChatGPT possibilities. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

31 pages, 6312 KB

Open AccessArticle

An Explainable Student Performance Prediction Method Based on Dual-Level Progressive Classification Belief Rule Base

by Jiahao Mai, Fanxu Wei, Wei He, Haolan Huang and Hailong Zhu

Electronics 2024, 13(22), 4358; https://doi.org/10.3390/electronics13224358 - 6 Nov 2024

Cited by 12 | Viewed by 4022

Abstract

Explainable artificial intelligence (XAI) is crucial in education for making educational technologies more transparent and trustworthy. In the domain of student performance prediction, both the results and the processes need to be recognized by experts, making the requirement for explainability very high. The [...] Read more.

Explainable artificial intelligence (XAI) is crucial in education for making educational technologies more transparent and trustworthy. In the domain of student performance prediction, both the results and the processes need to be recognized by experts, making the requirement for explainability very high. The belief rule base (BRB) is a hybrid-driven method for modeling complex systems that integrates expert knowledge with transparent reasoning processes, thus providing good explainability. However, class imbalances in student grades often lead models to ignore minority samples, resulting in inaccurate assessments. Additionally, BRB models face the challenge of losing explainability during the optimization process. Therefore, an explainable student performance prediction method based on dual-level progressive classification BRB (DLBRB-i) has been proposed. Principal component regression (PCR) is used to select key features, and models are constructed based on selected metrics. The BRB’s first layer classifies data broadly, while the second layer refines these classifications for accuracy. By incorporating explainability constraints into the population-based covariance matrix adaptation evolution strategy (P-CMA-ES) optimization process, the explainability of the model is ensured effectively. Finally, empirical analysis using real datasets validates the diagnostic accuracy and explainability of the DLBRB-i model. Full article

(This article belongs to the Topic Explainable AI in Education)

► Show Figures

Figure 1

Topic Menu

Topic Editors

Explainable AI in Education

Topic Information

Keywords

Participating Journals

Published Papers (12 papers)

Further Information

Guidelines

MDPI Initiatives

Follow MDPI