Question generation method and device, electronic equipment and storage medium

Abstract

The invention provides a question generation method and device, electronic equipment and a storage medium, and the method comprises the steps: determining a to-be-processed knowledge sub-graph from a knowledge graph, the knowledge graph being obtained by carrying out graph structuring on a knowledge text, and part of nodes in the knowledge sub-graph carrying answer node marks; the knowledge sub-graphs are input into a question generation model, a question text output by the question generation model is obtained, the question generation model is obtained through training based on a sample question text and a corresponding sample knowledge sub-graph, part of nodes in the sample knowledge sub-graph carry sample answer node labels, and a knowledge graph of non-strict ontology constraint is constructed; the domain correlation between the knowledge text and the knowledge graph is ensured, the knowledge modeling fineness is improved, the influence of noise in the knowledge text is reduced, the defect that in a traditional scheme, a question and answer pair generation method is not suitable for a long knowledge text is overcome, the utilization rate of structural information of the long knowledge text is improved, and the user experience is improved. The problem text is generated.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a question generation method, device, electronic equipment and storage medium. Background technique [0002] The question-and-answer knowledge base is the data foundation for building an intelligent question-answering system, and the question-and-answer knowledge points are the benchmarks for constructing the question-and-answer knowledge base. Most of the question-and-answer knowledge points are presented in the form of question-and-answer pairs. In scenarios such as intelligent customer service and business training, it is usually necessary to sort out the established question-and-answer knowledge points in advance. Such question-and-answer knowledge points are largely derived from long unstructured texts in the corresponding field, such as product manuals, insurance agreements, etc. etc., which makes the manual process of combing and summarizing the kno...

AI Technical Summary

Problems solved by technology

[0004] The present invention provides a question generation method, device, electronic equipment and storage medium to solve the defect that the question-answer pair generation method in the prior art cannot be applied to long knowledge texts

Images