Movie recommendation method and system based on multi-modal knowledge graph, and terminal

Abstract

The invention discloses a movie recommendation method and system based on a multi-modal knowledge graph, and a terminal and the method comprises the steps: obtaining movie pictures and text information according to single-modal features, extracting entities, attributes and relationships in the movie pictures and the text information, after obtaining the knowledge ontology, performing knowledge fusion and knowledge processing on the knowledge ontology and knowledge graph structured data to obtain a multi-modal knowledge graph; projecting the head entity and the tail entity into a vector space through an embedding layer, mapping the head entity and the tail entity into two embedded vectors, and projecting each relation in the path as an independent embedded vector; utilizing a long short-term memory network model to learn an aggregation path of embedded vector output representation entities and relationships; and projecting the final state of the aggregation path by using a full connection layer, outputting path scores, aggregating all the path scores through average pooling, outputting a prediction probability of watching a movie by the user, and displaying the prediction probability as a recommendation score to the user. Comprehensiveness and accuracy of movie recommendation are improved.

Description

technical field [0001] The present invention relates to the technical field of movie recommendation, in particular to a movie recommendation method, system and terminal based on a multimodal knowledge map. Background technique [0002] With the vigorous development of the Internet, the data content on the network is growing at an explosive rate. The simultaneous presentation of excessive information makes it difficult for users to obtain useful parts from it, and the efficiency of information use decreases instead. Recommendation algorithm has become a key technology for information data extraction and content mining. By combining the user's historical movie click records, in the face of a huge amount of information data, efficiently locate satisfactory recommendation information for users, which is the most effective way to improve the competitiveness of enterprises and an effective way to generate higher profits. Correspondingly, how consumers can quickly find the target ...

AI Technical Summary

Problems solved by technology

[0009] The main purpose of the present invention is to provide a movie recommendation method, system and terminal based on a multi-modal knowledge graph, aiming at solving the prob

Images