Research And Implemen Tation Of Online Learning System Based On Personalized Recommendation

With the popularity of the online lifestyle and the innovation of learning style,more and more online learning systems have emerged,allowing users to learn in the system at anytime and anywhere.Not only can it provide users with convenience,but bring troubles.It's not easy for users to find the resources they are interested in from the tremendous amount of learning resources quickly.Therefore,this paper applies a personalized recommendation technology to the online learning system to help users find the resources they are interested in from the tremendous amount of learning resources,which can recommend to users.Firstly,we analyze the key technologies used in this paper.According to the problems in the current recommendation algorithm of learning resources,a probabilistic matrix factorization personalized recommendation model integrating with deep learning is proposed.Through the Attention-CNN network and LSTM network we can get the course latent feature vectors and user latent feature vectors more accurately,which improve the accuracy of recommendation.At the same time,cluster and store the latent feature vectors obtained by the model to provide data basis for the design of course recommendation engine.Secondly,based on the clustering results of latent feature vectors and course information data,a course recommendation engine is designed.The engine can calculate and generate the recommendation candidate sets of various recommendation scenarios,including personalized recommendation candidate sets,related course recommendation candidate sets and popular course recommendation candidate sets,which can provide data support for the course recommendation.Then,according to the above model,course recommendation engine and specific business requirements,a low-coupling,easy-to-expand,high-availability and high-performance system architecture is designed for the online learning system and the design schemes of all kinds of server modules in the system are provided.Then,we use SpringBoot+SpringMVC + Mybatis + SpringCloud development framework to achieve the various modules of the system.Finally,we make a test both in the function and performance of the online learning system.The test results show that the system reaches the expected goals both in function and performance.