Design And Implementation Of Learning Resource Recommendation System Based On Knowledge Graph

With the development of Internet technology,the way people acquire knowledge is also quietly changing,from offline to offline and online acquisition,and online acquisition is the main method.At present,the online learning community and its related applications are developing rapidly,and people generate a large amount of original data related to interactive behaviors in the learning process,such as likes,comments,etc.In the era of widespread application of big data,how to process and utilize these raw data has become a research hotspot.People try to discover potential patterns from massive raw data,and push appropriate and customized learning resources for learners according to the behavioral characteristics of users.It can not only improve the user’s learning experience,but also improve the user’s learning efficiency,recognition and viscosity.This project uses the important demonstrations of learners to create knowledge maps,and provides learning materials suitable for the learners of this project with the following personalized learning resources.1.Build an online learning community knowledge graph.Collect and organize learners’ interactive behavior data,use named entity recognition and relation extraction to construct learners’ social network and learner knowledge acquisition network respectively,form a knowledge graph of online learning community,and vectorize the graph to obtain entities vector.2.Design a recommendation algorithm based on trust and similarity between learners.Using the weight and entity vector of the relationship between entities in the knowledge graph of online learning community,the trust degree and similarity between learners are calculated,which is used as the basis for the system to recommend learning resources to learners,which alleviates the problem of data sparsity.3.Design and build an online learning community recommendation system.The system includes the front-end interaction module of the system and the back-end algorithm engine module.The front-end interaction module of the system is developed using the Flask framework,and is divided into front-end display sub-modules and business logic sub-modules.The back-end algorithm engine module is divided into a graph data processing sub-module,a recommendation model training sub-module and a recommendation list generation sub-module.At present,the knowledge graph of the online learning community has been successfully constructed,and Neo4 J is used for storage and visualization,including718,961 triples,which are vectorized to obtain a 200-dimensional entity vector.The experiment of the learning resource recommendation system based on knowledge graph shows that the recommendation algorithm based on the trust and similarity between learners is better than the traditional recommendation algorithm,and the recommendation quality is improved.