A Course Knowledge Analysis Framework for On-line Education

Online education has been widely welcomed by learners for its convenience and rich interactivity. However, when facing the huge number of online courses, learners will have difficulty in choosing. The knowledge points contained in the course can effectively reflect the main teaching tasks of the course, making it easy for learners to quickly select courses they need. In this paper, we propose a random walk-based system for Courses knowledge analysis in a tag-knowledge bipartite network. First, we use TextRank to extract keywords from course texts as tags to describe the knowledge points according to annotated data. Next, tag-knowledge bipartite network is constructed by using the tags and knowledge points as nodes and the descriptive relationships between them as edges. Finally, we use Random walk to measure the relevance score between courses and knowledge points then return the top k relevant knowledge points. Experiments on real data sets have demonstrated the effectiveness and accuracy of the system.