Knowledge graph data extension method based on symmetric and reciprocal relationship statistics

Abstract

The invention discloses a knowledge graph data expansion method based on symmetric and reciprocal relationship statistics. The method comprises the following steps: firstly, performing duplicate checking deletion and information desensitization simplification, and compiling a dictionary; secondly, storing the reliable paired number of triples of the symmetric relation and the reciprocal relation through a two-dimensional matrix, and calculating the percentage of the triples meeting the special relation to obtain a special relation matrix table; then, setting effective thresholds, identifying relationships (relationship pairs) with percentages exceeding the thresholds as fully symmetrical (reciprocal) relationships, and creating unpaired triples under the relationships (relationship pairs)to extend the dataset. According to the method, a brand-new data expansion method based on statistical reasoning of the symmetric relation and the reciprocal relation is adopted, the number of data sets is directly and effectively increased, information hidden in the knowledge graph is mined, and the training effect of the knowledge graph representation learning model is improved.

Description

technical field [0001] The invention belongs to the field of knowledge graphs, and in particular relates to a knowledge graph data expansion method based on symmetric and reciprocal relationship statistics. Background technique [0002] The knowledge system is reorganized by human knowledge structure, such as WordNet language knowledge base, Freebase world knowledge base, etc. Knowledge base is an important basic technology to promote the development of artificial intelligence disciplines and support intelligent information service applications (such as intelligent search, intelligent question answering, personalized recommendation, etc.). The knowledge base mainly describes the relationship between entities in the real objective world. This knowledge is contained in the Internet information without (semi) structure, but the knowledge base is structured. Therefore, the main research goal of the knowledge base is to obtain structured knowledge from unstructured (semi-)struc...

AI Technical Summary

Problems solved by technology

At present, the vast majority of open domain knowledge graphs are incomplete in terms of content expression. When training the expression model, the lack of positive samples will directly affect the accuracy of knowledge representation.

Images