Learning Resource Recommendation Method Based on Knowledge Graph and Neural Network

In view of the traditional learning resource recommendation technology for different learners' learning style, knowledge level, learning mode and other characteristics of the difference, lack of personalized consideration, this paper proposes a individualization online learning resource recommendation method based on knowledge graph. Firstly, according to the knowledge structure and ability of students, the cognitive model of students is established, and the knowledge graph is used to describe the sequence relationship between knowledge points and learning resources; secondly, the word embedding technology is used to vectorize the information of learners' characteristics and learning behavior, and the learner characteristics are integrated and embedded into the recommendation model; finally, the bidirectional long short-term memory network is used Memory and attention mechanism are used for feature extraction to mine learners' implicit feedback information, so as to achieve the purpose of personalized recommendation for learners. The experimental results show that the proposed model is better than the traditional recommendation algorithm, and greatly improves the personalized learning efficiency of learners.