Personalized page-transition recommendation model based on contents and students' logs in digital textbooks
Zichen Kang, Chengjiu Yin · 2023
The e-book systems have been widely used in educational activities, such as college studies. At the same time, a huge amount of data regarding the learning history are collected by using the E-book system. Data mining and machine learning method have been used for analyzing these educational data in order to study the learning behaviors of students. Most rereading behaviors to study relevant contents on other pages again are inefficient due to that students are difficult to accurately navigate to the desired page with just a one-time page transition on the e-book system. In this study, we develop a hybrid recommendation model, which can recommend students to find the desired pages when rereading. The developed model is combining a content-based method (TF-IDF model) and a data-driven method (Page jumping model) linearly. By optimizing its parameters with the collected learning data, the developed model can generate ranking lists of relevant pages for students. Finally, the performance of the developed model is evaluated by two metrics: precision and recall.