Research On Online Course Recommendation Algorithm And The Knowledge Point Association Mining Method

With rapid the growth of information technology,education informatization has always been an important direction of education reform.As an important approach of education informatization,online education not only makes learming activities convenient and efficient,but also brings a large number of learning behavior data.Mining the information hidden in the learning behavior data can be used to improve the online education platform and optimize the students' learning process.Based on the "xAPI-based LMS platform" projecte,this paper designs the cloud architecture scheme of LMS platform,introduces the acquisition and storage of learning behavior data,emphasis on the data mining module of LMS platform,and analyzes the collected learning behavior data from the perspective of online course recommendation and knowledge point association.In addition,this paper presents an online course recommendation algorithm and knowledge point association mining method.The main work of this paper is summarized as follows:1.Research on the the xAPI-based LMS platform:First,this paper presents the cloud server deployment program which enhance the stability and availability of LMS platform.And through the message queue,this paper achieves the decoupling and asynchronous processing for LRS system and LMS platform.Secondly,combined with the online course organizational structure,this paper improves the courseware packaging method.In the study of learning behavior mining,LMS platform implements the data mining module based on Mahout.2.Research on algorithm of online course recommendation:First of all,this article combined with the user's feedback behavior and the characteristics of online courses to achieve the user interest model.Then,by improving the pheromone update strategy and the calculation method of the heuristic value,an algorithm based on ant colony algorithm is proposed.Experiments show that compared with the traditional recommendation algorithm,the proposed algorithm is more accurate and has obvious advantages in the recommendation of the course.3.Research on knowledge point association mining method:First,this paper improves method of building FP-tree,Proposes the knowledge mining algorithm based on multi-support strategy,and optimizes the processing method of modifying minimum support threshold.Then,based on the feedback information of the students for knowledge point association,the method of amending the knowledge point association is put forward,and the recomnendation method based on the knowledge point association is given.Experiments show that the knowledge point association mining method proposed in this paper can effectively mine the association between knowledge points.Compared with FP-growth algorithm,the knowledge points' recommendation based on this algorithm have obvious advantages in precision and coverage.