A Page Jump Recommendation Model Based on Digital Textbook Contents and Student Log Data

Abstract

In the analysis of student learning data, researchers found that students often switch pages but cannot accurately find the target page. Therefore, in this study, a page recommendation model to provide students with the target page for jumping to improve the efficiency of student page jumping behavior was developed. The recommendation model linearly combines the content-based recommendation model using term frequency-inverse document frequency (TF-IDF) and the recommendation model based on student log data, and determines the optimal weight of linear combination through gradient descent. Finally, the accuracy of the model was evaluated by comparing it with expert-generated recommendation outcomes.