Design of Recommendation Algorithm Based on Knowledge Graph

Abstract The commodity data in the shopping system contains a wealth of feature information, and different commodities are related by these features. Traditional collaborative filtering algorithms represent structured way, but they have issues such as low similarity calculation accuracy, poor recommendation effect, unfriendly results. algorithm based on knowledge graph put forward this paper initially automatically extracts entities entity relationships vertices edges then stores Neo4j database. Finally, between is calculated graph’s path similarity, list TOP-K with highest chosen for recommendation. experimental findings indicate that can not only more accurately commodities, also structure display recommended user-friendly manner, has better explainability well higher trust.