Learning Behavioral Pattern Analysis Based on Digital Textbook Reading Logs.

Since various features have different degrees of association with learning outcome, it is necessary to evaluate each feature by giving a reasonable weight. In this paper, we propose a different weighting of the features, weighting the features is different from other researches, when we carry out Correlation Analysis between these features and grade achieved (in the end of the semester). By using this weighting of the features and the students’ grade achieved, we grouped students into five clusters and analyzed their learning behavioral patterns. We found some interesting patterns, such as the students who always use the MEMO or MAKERS function, always get better grade achieved.