A knowledge map representation learning method based on entity and relational structure information

Abstract

The invention provides a knowledge map representation learning method based on entity and relationship structure information, the method comprising the following steps: acquiring structural semantic information of the entity and structural semantic information of the relationship in the knowledge map; according to the structural semantic information of the entity and the structural semantic information of the relationship, constructing an entity target vector and a target relation vector; constructing a score function according to the entity target vector and the target relation vector; constructing a loss function according to the score function, and learning an optimal vector representation of an entity and a relationship by minimizing the loss function. The present invention makes fulluse of structural information around entities and relationships to enrich and constrain the representation of entities and relationships. The method effectively enhances the expression ability of theentity and the relationship, constructs a new objective function, better expresses the entity and the relationship, and preserves the relationship between the entity and the relationship, so that themethod can be well applied to the large-scale knowledge map complement.

Description

technical field [0001] The invention relates to the field of knowledge graph natural language processing, in particular to a knowledge graph representation learning method based on entity and relationship structure information. Background technique [0002] With the advent of the era of big data, knowledge graph has become a current research hotspot. The emergence of knowledge graph is the inevitable result of artificial intelligence's demand for knowledge. Of course, its development is the result of the joint development of different research fields, not in the same strain. The knowledge graph itself is a network knowledge base formed by linking entities with attributes through relationships. The research value of the knowledge map lies in the fact that with the help of the knowledge map, the connection relationship between concepts can be established on the Web page, so that the information in the Internet can be organized at the minimum cost and become knowledge that can...

AI Technical Summary

Problems solved by technology

[0005] In view of the shortcomings of the prior art described above, the purpose of the present invention is to provide a knowledge map representation learning method based on entity and relational structural information, to solve the problem of not fully considering the rich structural semantic information of entities and relations in the prior art, As a result, the vector representation of entities and relationships cannot contain rich structural semantic information.

Images