Knowledge graph complex relationship reasoning method based on multidirectional semantics

Abstract

The invention provides a knowledge graph complex relationship inference method based on multidirectional semantics. The method comprises the steps of mapping entities in a training sample data set of a knowledge graph into two groups of low-dimensional space vectors for representation; mapping a relationship in a training sample data set of the knowledge graph into two groups of low-dimensional space vectors and one-dimensional parameter representations; entities in the training sample data set of the knowledge graph are randomly selected to replace entities of the training sample positive triad to generate training negative sample data; defining an objective function in the training process according to the training sample positive triad and the generated training negative sample; respectively substituting an entity mapping result and a relationship mapping result in the training sample data set into the target function, and optimizing to obtain vector representation corresponding to each entity or relationship in the knowledge graph; and calculating a distance value between the entity and the relationship in the knowledge graph triad by utilizing the vector representation obtained by optimization, and performing relationship reasoning according to the distance value. According to the method, the reasoning effect on the complex relationship is improved.

Description

technical field [0001] The invention relates to the technical field of knowledge graphs in artificial intelligence, in particular to a method for reasoning complex relationships in knowledge graphs by using an artificial intelligence representation learning method. Background technique [0002] With the development of artificial intelligence technology, knowledge graphs have attracted more and more attention from academia and industry, and knowledge graphs will play a pivotal role in the future development of artificial intelligence. The knowledge map is based on the triplet composed of head entity, tail entity and the relationship between them. The entity can be an entity in the real world, such as a specific person name, place name, institution, etc., and can also represent the attribute of an attribute Values ​​or concepts, such as a certain color, etc., the relationship can be the actual relationship between two entities and entities, such as husband and wife relationshi...

AI Technical Summary

Problems solved by technology

Although these models have shown their advantages and innovations in some aspects, these models only consider the semantic impact of relations pushing entities when using semantic information for complex relational reasoning, and do not consider the semantic impact of entities on entities and the semantics of entities on relations influences

For example (human, eat, vegetable) when learning the semantic representation vector, the semantic information of "human" should be affected by the semantic information of "vegetable" and "eat". Similarly, the semantic information of "vegetable" should also be affected by the semantic information of "human" and "eat". ", so that the learned semantic vectors represent insufficient semantic information of entities or relationships, which ultimately affects the reasoning effect on complex relationships

Images