Research On Movie Recommendation Based On Knowledge Map

With the rapid development of science and technology,the scale of data is also expanding,and we are exposed to massive amounts of data every day.In this context,the efficient acquisition of people’s information has also become an important issue.Traditional content distribution and search have been unable to fully meet people’s needs for information acquisition.In this context,personalized recommendation has become a hot spot at the moment.Personalized recommendation technology can proactively recommend information that may be of interest to users,reduce the cost of information acquisition,and improve the efficiency of information acquisition.Collaborative filtering algorithm is one of the current mainstream recommendation algorithms and is widely used.However,when faced with data sparsity and cold-start problems,the recommendation performance of collaborative filtering algorithms degrades.With the development of knowledge graph technology,this technology has been widely used in recommendation,search,intelligent question answering and other fields.The knowledge graph contains rich semantic information,and the hybrid recommendation algorithm combined with algorithms such as collaborative filtering can effectively solve the problem of data sparse and cold start,thereby improving the performance of the recommendation algorithm.The main research contents of this study include:The construction of knowledge graph in the film domain.Firstly,the movie dataset is supplemented,the movie entity and relationship type are determined,the entity table and relationship table are formed,the data is imported into the Neo4j graph database,and finally the knowledge graph is established.Knowledge graph feature learning and semantic similarity calculation.Based on the knowledge representation learning method,the entities and relations in the movie knowledge graph are embedded into low-dimensional vectors,and the semantic similarity between movies is calculated.Collaborative filtering recommendation algorithm based on knowledge graph.The rich semantic information in the knowledge graph is used as auxiliary information and integrated into the collaborative filtering algorithm for similarity calculation.Find a better fusion factor and verify it.