<![CDATA[The development of e-learning systems has generated a vast volume of user interaction data. Every activity—such as logging in, viewing materials, taking quizzes, and downloading assignments—contains valuable information that can be leveraged to enhance the effectiveness of online learning systems. This study aims to analyze user interaction association patterns in an e-learning system using the Apriori algorithm. A data mining approach was employed to identify relationships among features frequently accessed together, with a minimum support threshold of 0.4, minimum confidence of 0.6, and lift > 1.0. The dataset used consists of simulated (dummy) data representing seven user transactions and five main e-learning features. The analysis produced eight significant association rules with lift values above 1.0, indicating non-random relationships among features. Feature combinations such as {login} → {view_material} and {take_quiz} → {view_score} exhibited strong relationships, with confidence values reaching 0.75. These findings suggest the existence of dominant user interaction patterns that can be utilized to optimize navigation design, recommendation features, and overall user experience in e-learning systems. This research contributes to the application of the Apriori algorithm for exploring user access patterns in online education contexts, providing an analytical foundation for developing more adaptive and behavior-driven systems.]]>
Copyrights © 2025
