Knowledge graph completion method based on entity description and relationship path

Abstract

The invention relates to a knowledge graph completion method based on entity description and relationship path, which comprises the steps of S1, establishing a continuous bag-of-words model based on entity text description, and performing vector representation on entity description of entities in a knowledge graph by utilizing the continuous bag-of-words model to obtain a description-based vector;S2, according to a conversion-based model between the entity vector and the relation vector and between the entity vector and the relation path, obtaining a conversion-based model, establishing a score function of the relation triad (h, r, t) and the path triad (h, p, t) and a loss function of the score function, learning vector representation of an entity, a relation and a path by minimizing theloss function, and learning vector representation based on a structure; and S3, adopting the learned entity vector representation, obtaining representation results in the vector space in different tasks, and complementing the knowledge graph or mining the potential relationship.

Description

technical field [0001] The invention relates to a knowledge graph completion method based on entity description and relationship path, and belongs to the technical field of the Internet. Background technique [0002] In recent years, with the rapid development of Internet technology and application models, it has triggered an explosive growth in the scale of Internet data, which contains a large amount of valuable knowledge. How to organize and express this knowledge, and how to conduct in-depth calculation and analysis on it has attracted much attention. Knowledge graphs emerged as a rich and intuitive way of expressing knowledge. Knowledge reasoning for knowledge graphs is one of the research hotspots of knowledge graphs, and has played an important role in vertical search, intelligent question answering and other application fields. Knowledge reasoning based on knowledge graph aims to infer new knowledge or identify wrong knowledge based on existing knowledge. Differen...

AI Technical Summary

Problems solved by technology

However, the existing knowledge graph representation learning methods need to learn the structural relationship between entities during training, while ignoring additional information such as entity text descriptions in the knowledge graph, knowledge graph completion and entity classification tasks such as accuracy comparison Low, the existing knowledge map completion algorithm based on translation transformation can capture the inherent characteristics of the data well, the algorithm is efficient and easy to operate, but the algorithm relies on pure data-driven, and can predict the relationship between entities. Good explanation, but severe data sparsity problem occurs in low connected graphs

Images