Subproject 1: Applying Big Data Technique to Moocs Learners’ Course Video Analytics and Adaptive Course Material Recommendation(2/3)

MOOCs can keep students’ learning records of the leaning activities they join. By big data analysis, students’ learning conditions and learning behaviors can be predicted from their learning records. Furthermore, the model can be built and the hidden learning mode can be find out so that the instructor can know whether students face learning difficulties or they need extra help. The main method to achieve the aforementioned goals is the big data technology.Owing that the teaching materials of MOOCs are usually the learning films that are recorded in advanced, students may face the problem of not having enough background knowledge of the course they take and not having enough assistance while they get into difficulties. Once students do not understand the course film, MOOCs system does not provide them with instant feedback mechanism. Students may not know how to overcome their problems. Likewise, the instructor may not know how to help students because MOOCs system does not have student’ feedback. To solve the aforementioned problems, sub-project 1 provides correlation analysis of course films and the recommendation model of adaptive teaching materials. This model will recommend teaching materials according to the best learning path provided by the main project and the course contents. In addition, combining with news and events, the recommendation model enriches the adaptive teaching materials. In addition to the in-class knowledge, Students can learn professional knowledge outside the classroom by the recommendation model. Sub-project 1 also provides course film navigation to help students realize the film and further absorb the knowledge. Moreover, sub-project 1 will recommend students with adaptive materials according to the materials suggested by the best learning path in the main project, the learning process analysis and the learning behavior analysis.