Learning Path Recommendation System Based On Knowledge Graph

With the rapid development of Internet education,online learning is becoming more and more popular.Learners can get massive learning resources from the Internet or choose online learning platform for independent learning.At the same time,learners can also freely choose learning time,learning content.In the face of this highly autonomous learning method,it is difficult for learners to accurately recognize the existing knowledge state,and clearly understand the current learning progress and cognitive compliance state.Therefore,it is extremely easy to fall into problems such as learning loss,lack of learning motivation and so on.Based on this situation,personalized learning path recommendation can effectively solve the problems of knowledge loss,cognitive overload and lack of learning motivation caused by online learning.At present,personalized learning path recommendation algorithms mainly include knowledge-based recommendation,contentbased recommendation,collaborative algorithm recommendation and hybrid recommendation.Among them,collaborative filtering recommendation algorithm is widely used.However,the "Matthew effect" of collaborative algorithm in the field of personalized path recommendation affects the learning effect of learners with low cognitive level.In view of the above problems,this paper proposes a personalized learning path recommendation model based on three-way decision.The model takes the educational knowledge graph as the bottom recommendation content,combined with the three-way recommendation methods based on cognitive level,to recommend the learning path content for learners.In view of the current problems of mixed e-learning resources,to ensure the quality of the recommended content of the follow-up learning path,this paper constructs the Python educational knowledge graph.In the research of constructing knowledge graph,this paper mainly involves two parts.The one is to construct the pattern ontology of knowledge graph,which is to semantically express the concepts,attributes,and relationships in the field of education by using the ontology library Protégé.And the other is to use Neo4 j for knowledge storage.Through Neo4 j storage and visual display,it can not only improve the solutions for online learners to learn the problem of trek,but also provide support for personalized learning path recommendation.Aiming at the common problem of "Matthew effect" in the field of personalized path recommendation,this paper proposes a personalized learning path recommendation model based on three-way decision.The model can be divided into two parts: user model and personalized learning path recommendation model.Firstly,through the analysis of online users' learning behavior data,the dimensions and measurement indicators of online learners' learning model are constructed to describe online learners' learning behavior habits,learning preferences,and learning level.Then,the personalized learning path recommendation algorithm based on three-way decision is used for learning path recommendation.Because different learners have different mastery of knowledge and the ability to accept new knowledge,they are divided into three branches according to learners' cognitive level,and then put forward different strategies for each part of learners and recommend personalized paths suitable for their cognitive level.The results show that this path recommendation can effectively alleviate the problems of weak learning motivation,knowledge loss and cognitive overload of online learners in the e-learning.