On Using Data Mining For Browsing Log AnalysisIn Learning Environments

Recently, the rapid progress of Internet technology has triggered the widespread development of web-based learning environments in the educational world. As compared with conventional CAI systems, web-based learning environments are able to accumulate a huge amount of learning data. As a result, there is an urgent need for analyzing methods of discovering useful knowledge from the huge log database for improving instructional/learning performance. In this chapter, I will present some models and methods of analyzing the browsing log data to construct a browsing behavioral model that is helpful in supporting e-learning applications. For example, teachers can investigate the model to identify some interesting or unexpected learning patterns in student’s browsing behavior, which might therefore provide knowledge for teachers to reorganize their content structure in a more effective manner. Alternatively, another model can be used as a reference model by which personalized content recommendation could be made. To serve these purposes, a set of tools based on data mining techniques such as clustering, and association mining, combined with collaborative filtering techniques, are developed. The effectiveness of these methods is investigated on a real database collected from web-based courses. Through the case studies, some revelations are presented and some future research directions are discussed.