Online Education User Behaviour Based on Machine Learning

Abstract

The rapid development of machine learning has made it possible to quickly mine exploitable data from massive amounts of data, and it is important for online education platforms to reasonably track the changing learning behaviour of learners. The aim of this paper is to study the prediction of online education user behaviour based on machine learning. A deep clustering algorithm is proposed based on demographic information about learning behaviour and clickstream data recorded in a university virtual learning environment. Applying the deep clustering model proposed in this paper to the analysis of the distribution of learning activities and the correlation between learning behaviour and education level, it is demonstrated that the deep clustering model proposed in this paper's can analyse and predict the learning behaviour of different student groups. It helps us to further study and analyse students' learning behaviour in depth and comprehensively in order to provide real-time quality feedback and promote the development of online education.