Similar character recommendation method based on knowledge graph representation learning

Abstract

The invention relates to a similar character recommendation method based on knowledge graph representation learning, which adopts a brand-new design strategy and comprises social user information processing, representation learning model training, clustering index of character point vectors in entity relationship vectors and similar character recommendation, a TransD representation learning algorithm is adopted to perform knowledge representation learning on all entities and relations in the knowledge graph, and semantic information implied in the graph is mined more deeply; for large-scale figure point vectors, an Annoy rapid calculation distance algorithm is adopted to cluster similar figures, and as semantics in a graph are considered in entity relation vectors, powerful information support is provided for clustering; for the incremental updating condition of the social knowledge graph, a representation learning recommendation algorithm is designed for periodic training, and the freshness of a recommendation result is kept; the whole design method can effectively improve the character similarity detection precision and obtain a more accurate similar character recommendation effect.

Description

technical field [0001] The invention relates to a method for recommending similar characters based on knowledge graph representation learning, and belongs to the technical field of knowledge graph representation learning. Background technique [0002] With the rise of mobile Internet, social network has become a typical product of mobile Internet, which strengthens the close communication between people, and character recommendation also plays a vital role in social network. [0003] There are some character recommendation methods in the prior art: [0004] 1. Network personalized recommendation method based on PageRank algorithm: obtain the friendship relationship between the group and its members from the web page configuration file, and establish the personal preference model of each group member. The PageRank algorithm is used to iteratively calculate the influence of group members on the group, so as to obtain the preference model of the entire group, and use this mode...

AI Technical Summary

Problems solved by technology

[0006] Nowadays, with the development of big data, the interpersonal network is getting bigger and bigger. The existing technology has low applicability for large-scale social networks. There are a lot of semantic information in the huge social network. The existing technology fails to capture the social network. Semantic connections in human beings, and the development of artificial intelligence has entered the stage of cognitive intelligence. Cognitive intelligence requires machines to learn to process human complex language and perform knowledge reasoning, which is very difficult for machines

Images