Design and Evaluation of an AI-Based E-Learning System for Graduate Education

Assessing Student Needs for AI-based Adaptive Learning in Higher Education

The study contributes user-derived design requirements that can guide the development of trustworthy and context-appropriate AI-supported learning platforms for undergraduate ICT students in programming-related courses at the two participating universities; broader generalization to other higher education fields requires further research.

The Role of AI in Enhancing Teaching–Learning Practices across Disciplines

Artificial Intelligence (AI) is transforming teaching–learning practices by enabling personalized, data-driven, and scalable educational environments aligned with Industry 4.0. AI-based tools such as adaptive learning systems, intelligent tutoring, automated assessment, simulations, and predictive analytics address key limitations of traditional education, including one-to-many instruction, delayed feedback, and limited learner engagement. Across disciplines engineering, healthcare, humanities, social sciences, and management—AI supports contextual and experiential learning through virtual labs, NLP-based feedback, decision-support systems, and immersive technologies. Beyond instruction, AI enhances institutional functions such as learner analytics, dropout prediction, curriculum optimization, and inclusive education. This paper reviews recent research and proposes an AI-Integrated Pedagogical Enhancement Model (AI-IPEM). Empirical findings indicate improvements in learning efficiency (22–45%), feedback turnaround time (70–90%), student retention (10–18%), and learning compliance (30–50%). The study concludes that AI serves as an enabler of augmented pedagogy, complementing teachers and fostering higher-order thinking, creativity, and lifelong learning.

A Web-Based Chatbot Intelligent Tutoring System to Support the Improvement of Higher-Order Thinking Skills (HOTS)

The improvement of technology has impacted significant changes to education, including mathematics learning, which often gives challenges, such as limited interaction and low student interest to learn. Advancements in artificial intelligence (AI) have created new way to improve learning effectiveness. Therefore, this study focused on the development of an Intelligent Tutoring System (ITS) based on a chatbot incorporated into a web to support higher-order thinking skills (HOTS). The ITS used the pre-trained GPT-3.5 Turbo model to provide students with relevant, interactive responses. This study aimed to create more engaging, personalized, and effective learning media to help students understand mathematical concepts. The analysis used a Research and Development (R&D) methodology with the ADDIE (Analysis, Design, Development, Implementation, and Evaluation) development model. During the process, the subjects of the media trial were 24 junior high school students and five teachers from SMPN 6 Kota Bima. The results showed that the developed web-based ITS had 89% validity level, 84.42% practicality, and a media effectiveness of 70.83%. The values signified that the media was considered valid, practical, and effective. This shows that this chatbot has the potential to be an innovative solution to improve HOTS.

AI-Driven Educational Transformation in Higher Education: A Review of Current Practices

Despite the fact that artificial intelligence has already gained ground in higher education systems, especially in the field of online learning, it is important not to mistake its development for actual transformation. In this section, we will show that AI is rapidly gaining a foothold in teaching, learning, and online learning. To verify the hypothesis concerning the positive impact of the use of AI in class, we will review the empirical data from the authoritative institution Paper, a test conducted in the course of professional training, positive results achieved through intelligent tutoring, learning analytics and automatic feedback, and supportive arguments regarding the answer in the view of prepared arguments. In comparing these statistics, there have been 45,398 students and faculty respondents from 35 countries surveyed for the Digital Education Council (DEC) in the 2026 survey, 1,054 UK undergraduate students included in the 2026-HEPI survey, 400 higher education respondents responding from 90 countries surveyed by UNESCO in 2025, and 1,681 faculty included from 28 countries in the 2025 DEC-WG Faculty survey. Among the findings is an asymmetry of transformation: AI deployment activities are commonly observed; however, only 15% of students claim integration in many subjects, and only 29% rated their classroom teachers very highly in terms of abilities in guiding AI use, and only 28% rated most or many of the assessments as being appropriate to the use of AI in the workplace. An experiment within a physics course has validated that effective AI tutoring structures can be more effective than rudimentary active learning, but it has also been shown that most other AI classroom applications are primarily incidental and only indirectly related to educational design. It further argues that changes that last over time, or sustainable change, require coherent improvement of teaching, assessment, educator capabilities, leadership, and equity objectives and offers the ALIGN framework (Access, Learning design, Integrity, Governance, and new capabilities) as an evidence-based institutional implementation model.

Design and exploratory evaluation of an AI-supported blended learning environment for big data education

EduVa: Prototyping and testing AI-powered interactive LMS with adaptive modules and assessments

The findings demonstrate that the EduVa LMS is valid, usable, and effective for implementing adaptive learning, and shows a strong correlation between AI assessments and learning objectives.