Knowledge complementing method and device for knowledge graph

Abstract

The invention provides a knowledge complementing method and device for a knowledge graph, which can solve the problems that negative examples are meaningless and the 1-N relation is difficult to process in the whole knowledge complementation process.. The method comprises the steps of determining a space vector corresponding to an entity and a relationship; calculating a semantic relationship according to the entities and the space vectors corresponding to the relationships to obtain a new relationship between the entities, and complementing the knowledge graph; randomly generating a negativeexample by using the generative adversarial network, and training a first knowledge representation model in combination with the derived fact triple; performing concept layering on the obtained fact triple, randomly selecting entities under the same sub-concept of the fact triple to construct a negative example, and training a second knowledge representation model by adopting a maximum interval method in combination with the derived fact triple; and taking the second knowledge representation model as a discriminator input of the first knowledge representation model, and optimizing the first knowledge representation model through the adversarial generative network to obtain a target knowledge representation model for knowledge completion. The invention relates to the field of knowledge engineering.

Description

technical field [0001] The present invention relates to the field of knowledge engineering, in particular to a knowledge complement method and device for a knowledge graph. Background technique [0002] Knowledge graphs are often represented in a highly structured form, describing the relationships between various entities in the real world. At present, knowledge graphs have been widely used in various fields, such as: automatic question answering, information extraction and other fields. A typical knowledge graph is composed of a large number of triples. Although knowledge graphs can provide high-quality structured data, most of the public knowledge graphs are constructed by manual or semi-automatic methods. These graphs often have the problem of data sparseness and even the relationship between a large number of entities is not fully understood. In order to obtain a higher-quality knowledge map, it is necessary to complete the knowledge map. [0003] The goal of knowled...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide a method and device for knowledge completion of knowledge graphs, so as to solve the problems in the prior art that the negative examples are meaningless and the 1-N relationship is difficult in the process of knowledge completion using the TransE model. Dealing with the problem

Images