Behavior Analysis in Distance Education by Boosting Algorithms

Student behavior analysis is an active research topic in distance education in recent years. In this article, we propose a new method called Boosting to investigate students’ behaviors. The Boosting Algorithm can be treated as a data mining method, trying to infer from a large amount of training data the essential factors and their relations that influence students’ academic successes. Based on the trained model, it is possible to predict students’ academic successes and to assist them to adjust their learning behaviors. Among the essential factors selected by the Boosting algorithm, the analysis and comparison are conducted between on-campus students and off-campus students. More importantly, these findings are of great importance to academic administrators, faculty members, and instructional developers in order to improve the teaching modes and online courseware design.