Journal Article

Using Data Mining Techniques in Tracking the Students' Behavior in the Asynchronous E-Learning Systems

Elawaj, Tareg Abdusalam, Algaet, Mustafa Almahdi, Adrugi, Salem Msaoud

2018

Overview

This applied research investigates the use of data mining techniques to analyze and track student behavior in asynchronous e-learning systems, with the goal of improving educational quality through data-driven insights. Conducted by Salem Msaoud Adrugia, Mustafa Almahdi Algaeta, and Tareg Abdusalam Elawajb, the study employs an empirical quantitative approach, extracting user logs from a university learning platform over one semester and applying algorithms such as K-Means Clustering, Decision Trees, and Association Rule Mining. The analysis reveals distinct behavioral patterns among learners, identifying three main clusters: \"active\" students who regularly engage and meet deadlines, \"delayed\" learners with sporadic access and reduced achievement, and \"passive browsers\" who read content without interaction. The study finds a strong positive correlation between consistent engagement and academic performance, and highlights behavioral metrics-such as message count, login frequency, and session duration-as potential early indicators of academic risk. The authors propose integrating data mining into academic decision-support systems to enable educators to monitor student participation and implement personalized interventions. They conclude that applying data analytics to e-learning environments represents a key step toward AI-driven education, where learning platforms evolve into intelligent systems capable of understanding and responding adaptively to individual learner behavior, thus fostering a more efficient and personalized educational experience. Abstract Written by Dar AlMandumh, 2025, Using AI

• Publisher: جامعة المرقب - كلية التربية بالخمس
• Subject: استخراج البيانات / التعليم الإلكتروني / التنقيب في البيانات / سلوك الطلاب