Mining User´s Interest from Reading Behavior in E-learning System

A method based on reading behavior is proposed for capturing user´s interest in E-learning System. A topic ontology is necessary to be predefined which can be used as the reference to construct user´s interest model. A behavior table is generated by profile agent to record user´s behaviors including underline, highlight, circle, annotation and bookmark. We adopt a behavior matrix and weight matrix to compute the user´s interests in each leaf topic in the topic ontology. The user´s interest about the topic ontology can be extended via support factor. An experiment is conducted to estimate the performance of our approach. It shows that our approach can capture user´s interest precisely.