A MULTI-AGENT MACHINE LEARNING FRAMEWORK FOR PERSONALIZED LEARNING STYLE CLASSIFICATION IN STEM EDUCATION

Abstract

Background: In educational settings, the diversity of individual learning styles often remains under-addressed, leading to variable comprehension and engagement among students. Traditional teaching approaches may not fully accommodate the distinct needs of each learner, particularly in STEM education where personalized, adaptive instruction can greatly benefit understanding and retention. Objective: This study aims to develop an intelligent multi-agent framework that can accurately identify and classify learning styles—Visual, Auditory, and Kinesthetic—through a STEM-focused approach. The objective is to create a system that personalizes educational content by analyzing learners’ feedback and engagement, thereby enhancing learning outcomes. Method: The proposed framework employs a multi-agent system that includes a Teacher Agent, Concept Mapping Agent, Content Analysis Agent, and Sentiment Analysis Agent. Using feedback from learners, the system classifies each learner’s style by employing machine learning (ML) algorithms, including Multinomial Naive Bayes, Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). The performance of each algorithm is measured through Precision, Recall, Accuracy, and F1-score to ensure reliable classification. Results: The framework demonstrated high accuracy about 98%, with Multinomial Naive Bayes achieving the best classification results among the algorithms. The inclusion of sentiment analysis provided further insights into learners' engagement levels, supporting a refined understanding of each student’s preferred learning style. Conclusion: This study presents a successful model for personalized learning that dynamically adapts to individual styles, promoting inclusivity and engagement in STEM education. Future work could enhance this framework by expanding to more learning styles and refining real-time adaptability, paving the way for a fully responsive, student-centered educational experience.