Knowledge graph entity alignment method and device

Abstract

The invention provides a knowledge graph entity alignment method and device. The method comprises the steps of obtaining data of two knowledge graphs to be fused; performing neighborhood aggregation entity representation learning on the data of the two knowledge maps to obtain entity representations of entities in the two knowledge maps; performing relation representation learning for enhancing entity semantics according to the entity representation, and modeling a relation between entities to obtain an entity relation representation; carrying out concept and concept hierarchical system representation learning according to the entity representation, and carrying out modeling on relationships between entities and concepts and between concepts to obtain concept and concept hierarchical system representation; constraining the entity representation in the vector distance-based entity alignment process through the entity relationship representation and the concept and concept hierarchy system representation, and obtaining an alignment result of the two knowledge graph entities. Concepts and a concept hierarchical system are fused into an entity alignment framework and are made to play a role, and the accuracy of entity alignment is improved.

Description

technical field [0001] The present invention relates to the field of natural language processing of computer artificial intelligence, in particular to a method and device for aligning knowledge map entities. Background technique [0002] Knowledge graphs that integrate multi-language and multi-knowledge source information have become an important source of knowledge for many artificial intelligence applications such as information extraction and intelligent question answering. In order to more efficiently fuse information in overlapping and complementary knowledge graphs, entity alignment has attracted the interest of many scholars and has become an important research problem. Many knowledge graphs provide rich structured knowledge for different applications. Due to the different construction purposes of these knowledge graphs, there is great heterogeneity among the knowledge, and some complementary knowledge will also be included. In order to better support tasks such as t...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a knowledge map entity alignment method and device, which are used to solve the defect of low entity alignment accuracy in the knowledge map in the prior art, realize the alignment of entities belonging to the same concept under the constraints of the concept, and improve the efficiency of entity alignment accuracy

Images