Query graph sorting method and system oriented to knowledge graph questions and answers

Abstract

The invention relates to a query graph sorting method and system oriented to knowledge graph questions and answers. The method comprises the following steps: analyzing natural language questions into corresponding candidate query graphs, and obtaining a candidate query graph set corresponding to each question, wherein the natural language questions comprise a training set, a test set and a verification set; selecting an optimal query graph from the candidate query graph set, including: converting the candidate query graphs into corresponding query graph sequences; constructing a sorting model; performing iterative training on the sorting model by using a training set to obtain a trained sorting model, evaluating the trained sorting model through a verification set, and storing the sorting model with the highest evaluation score as an optimal sorting model; and obtaining a score set corresponding to each question in the test set through the optimal sorting model, and screening the score set to obtain a query graph with the highest score. The coding quality of the query graph is improved, the sorting process is reasonable, and the overall performance of the system is improved.

Description

technical field [0001] The present invention relates to the technical field of natural language processing, in particular to a query graph sorting method and system for knowledge graph question answering. Background technique [0002] Knowledge graph question answering is one of the important tasks in the field of natural language processing. It takes natural language questions as input, and finds relevant entities or attribute values ​​from the knowledge graph as the answer output. The emergence of knowledge graph question answering provides a feasible solution for people to acquire knowledge quickly and accurately. At the same time, the continuous development of a variety of large-scale knowledge graphs has also laid the foundation for the application of knowledge graphs, making the question-and-answer application of knowledge graphs in the open field gradually enter the public's field of vision. [0003] Existing knowledge graph question answering methods can be divided ...

AI Technical Summary

Problems solved by technology

[0009] For this reason, the technical problem to be solved by the present invention is to overcome the prior art that ignores the interactive information between questions and query graphs, requires manual design of complex query graph features, and lacks the technical defects of considering the connection between candidates

Images