Design And Implementation Of A Course Ideological And Political Learning System Based On Hybrid Recommendation

Currently,China has made the construction of ideological and political courses,the improvement of students’ professional level and moral quality a key focus in promoting the development of higher education.However,with the rapid development of internet technology,the scale and quantity of various learning resources are constantly increasing.It is difficult for teachers and students to find course ideological and political resources related to the subject from the massive learning resources.Combining recommendation algorithms with course ideological and political learning systems can effectively solve this problem.The-collaborative-filtering algorithm and the content-based recommendation algorithm are the current mainstream recommendation algorithms.The former can recommend items for users only by combining users’ ratings,but there will be problems such as cold start and sparse rating matrix;The latter can solve the cold start problem by obtaining the feature attributes of users and items,but its scalability is poor.This thesis combines collaborative filtering algorithm and content-based recommendation algorithm to design a course ideological and political learning system based on hybrid recommendation.The main work is as follows:(1)Investigated the research status of the curriculum ideological and political learning system,proposed the research goal of the curriculum ideological and political learning system integrating collaborative filtering and content based recommendation algorithm,the theory of recommendation algorithm and the key technology of curriculum ideological and political learning system development.Taking the database course as an example,selecting corresponding ideological and political elements according to different chapters of the course,designing ideological and political teaching methods and content,laying the foundation for the subsequent implementation of ideological and political resource recommendation function.(2)Aiming at the problems of existing collaborative filtering recommendation algorithms and content-based recommendation algorithms,corresponding algorithm improvements are made.In Pearson similarity calculation,two penalty factor coefficients,popular resources and time decay,were introduced to solve the problem of inaccurate recommendation results caused by the influence of popular resources and time.Then,a content based recommendation algorithm was used to assign labels to resources,and the similarity between resources was calculated based on the weight of the labels.Finally,the similarity obtained from the two recommendation algorithms was combined to predict the user’s rating of the resources,Solved the sparse rating matrix and new user cold start issues,reduced errors,and improved recommendation accuracy.(3)Building a course ideological and political learning system based on Java language.Based on the requirement analysis,the functional modules,overall architecture,and database tables of the system were designed in detail.Combined with web development technology frameworks such as Springboot,Mybatis,Vue.js,and Uni-app,functions such as learning,uploading,searching,and personalized resource recommendation of course ideological and political resources were implemented.The system’s functionality was tested and verified.