Real-time Pattern Detection Methodology for Monitoring Student Behaviour on E-Learning Platform in the Field of Financial Sciences: Case Study

Abstract

Nowadays, we see that the students who are educated on an e-learning platform demand features that will increase the satisfaction of the learning experience. In order to meet these demands, e-learning platforms need a) systems that can monitor students' behavior in real time through clickstream data, and b) generate instant actions for assisting / directing / informing students. Within the scope of this research, a method that can produce instant actions according to the students' behaviors is proposed in order to increase satisfaction in student experience by using methods such as streaming data processing and analysis and complex event processing. In order to demonstrate the usability of the proposed method, a prototype application has been developed for an e-learning platform providing online lectures in the field of financial sciences. Tests were performed on the developed prototype application in terms of performance and scalability. The results show that the proposed method gives successful results in determining student behaviors in real time and taking actions.