Personalized Learning Model based on Deep Learning Algorithm for Student Behaviour Analytic

Abstract

The learning processes are unique for each individual. Thus, the appropriate learning methods are essential to boost the student performance and success in their study. The current e-learning system is focusing on producing tools for collaborative study or facilitating the course for the discussion forum and direct communication such as chat service. However, the problem of boosting up individual achievements remains challenging. This paper proposed a personalized learning model to find suitable learning methods based on a deep learning algorithm. The factors of personalized learning such as adaptive, individualized, differentiated and competence-based learning have been taken into account when build and tested the model. These factors will act as a feeder for the network then it will be processed to identify which model of learning has the biggest chance on augmenting student achievement. The testing result has shown that the proposed model with deep learning approach has overcome other classifier with 72% success rate The model is believed to bring a great benefit for student and the teacher as well to manage the studying process more convenient for both parties.