Knowledge graph construction method based on hierarchical category information

Abstract

Aiming at the defects of the existing knowledge graph construction method, the invention discloses a knowledge graph construction method based on hierarchical categories, which can utilize a Word2vectool and comprises the following steps of: labeling entity relationships, acquiring relationship category information, constructing hierarchical categories, hierarchically embedding knowledge graphs,and automatically adding hierarchical information into relationships of the knowledge graphs. The accuracy of knowledge graph modeling can be greatly improved, and the knowledge graph modeling methodcan be better applied to practical applications such as knowledge questioning and answering, intelligent searching and case matching.

Description

technical field [0001] The invention belongs to the technical field of knowledge graphs, and in particular relates to a method for constructing knowledge graphs based on hierarchical category information. Background technique [0002] Knowledge graph is essentially a large-scale semantic network, which contains various entities, concepts and semantic relationships. By representing knowledge in the form of a knowledge graph, knowledge can be better learned and expressed by machines. The embedded representation of the knowledge graph refers to representing the entities and relationships in the knowledge graph as a low-dimensional vector in the model, which further improves the computability of knowledge in the knowledge graph. [0003] After being trained by the knowledge spectrum graph embedding method in knowledge representation learning, the embedding results of entities and relations can be easily applied to downstream tasks or practical applications. For example: knowle...

AI Technical Summary

Problems solved by technology

[0004] Whether the knowledge graph can be effectively modeled has become the key to the application of the knowledge graph. The traditional method regards the relationship between entities as independent, ignoring many information hidden in the knowledge graph, such as the category of the relationship. , the correspondence between relationships, etc., so that only a limited model of the knowledge map can be constructed, and some details cannot be well described, which affects the accuracy of the application

Images