Research And Implementation Of Course Recommendation System For MOOC Platform

With the rapid development of the Internet and the growing demand for education,in 2012,a new online teaching model,called Massive Open Online Course(MOOC),emerged and spread rapidly around the world.Along with the rapid development of this large-scale open online course platform,the number of courses on the MOOC platform has increased significantly.The MOOC platform has an information overload problem similar to that of the e-commerce platform.Users will easily have trouble in choosing and handling the overloaded information when they encounter these problems of information overload.Therefore,the research and implementation of the course recommendation system for the MOOC platform has important practical application value.The main work of this paper is as follows:(1)Optimization of course recommendation model based on neural network.For the information overload problem,the traditional recommendation algorithm usually uses collaborative filtering.Due to thedevelopment of neural networks,many recommended models apply neural networks to collaborative filtering to improve the performance of the model.The course recommendation model constructed in this paper is improved on the basis of neural network-based collaborative filtering model.Considering the influence of social network on model recommendation,this paper integrates the user's concern relationship into the original model.In order to verify the improvement of the model by integrating the concern relationship,this paper used the actual data set from IMOOC to compare the experiment.The results show that the intergration of concerns relationship can make the original model perform better.(2)Second recommendation based on similar courses.In order to recommend courses on different MOOC platforms to users and meet the needs of users for course learning on different MOOC platforms,this paper proposes a second recommendation based on similar courses.Based on the TopK course recommended by the course recommendation model constructed in this paper,it can calculate more similar courses on the MOOC platform as the second recommendation of this K course.The similar courses in this paper are calculated by the full-text search server Solr on different MOOC platforms based on keyword search.In order to consider the quality of the second recommended courses,this paper re-sorts the retrieved course through integrating course scores and courseviewers into similar courses.Finally,the corresponding similar courses are recommended to the current users on different MOOC platforms to improve the breadth of the recommendation model.(3)Implementation of the course recommendation system.The course recommendation system implemented in this paper applies the course recommendation model constructed in this paper and the second recommendation based on similar courses proposed in the paper to the existing MOOC platform to provide users with personalized course recommendation through the form of system.