Natural language question answering method and system based on question sentence and knowledge graph structure analysis

Abstract

The invention discloses a question sentence and knowledge graph structure analysis-based natural-language question answering method and system, and belongs to the fields of graph calculation and natural-language question answering. For the existing problems of various question answering systems, the invention provides the question sentence and knowledge graph structure construction, analysis and matching-based natural-language question answering method, thus types of natural-language questions which can be answered are expanded, and accuracy of answers is improved at the same time. According to the method of the invention, a structure of a graph is adopted to represent a query question sentence through traversing centered on an entity node and design of an extraction rule; and mapping finding is carried out on sub-graphs, which cover all answers, in the knowledge graph on the basis of the query graph. Through constructing semantic vectors of paths and phrases, similarity of matching iscalculated while filtering and screening on the answer sub-graphs are realized, and thus candidate answers are obtained. According to the method of the invention, a solution is concise, and an effectis obvious.

Description

technical field [0001] The invention belongs to the fields of graph calculation and natural language question answering, and more specifically relates to a natural language question answering method and system based on question sentences and knowledge graph structure analysis. Background technique [0002] With the rapid development of the Internet, more and more large-scale structured knowledge graphs have been constructed. Knowledge graphs are playing an increasingly important role in integrating information and enhancing intelligent search. However, all the information stored in this knowledge graph is structured, which makes it difficult for ordinary users to use it directly. [0003] The query based on query language and data acquisition protocol (SPARQL Protocol and RDF Query Language, SPARQL) is a structured query for Resource Description Framework (RDF) data. Although the knowledge graph can be directly retrieved, it Users need to write corresponding query statemen...

AI Technical Summary

Problems solved by technology

[0006] In view of the above defects or improvement needs of the prior art, the present invention provides a natural language question answering method and system based on question sentences and knowledge graph structure analysis, thereby solving the problem of low query accuracy existing in existing knowledge graph query methods technical problem

Images