Method and device for decomposing complex questions in vertical field and computer equipment

Abstract

The invention discloses a vertical field complex question decomposition method and device and computer equipment, and the method comprises the steps: determining a vertical field to which a to-be-decomposed complex statement belongs, inputting the complex statement into a recognition model corresponding to the vertical field, and extracting a subject entity and an auxiliary word type; combining the subject entity and the auxiliary word type to form a feature expression, and matching the feature expression with a syntactic template library to generate a knowledge graph query statement; inputting the query statement into a knowledge graph corresponding to the vertical field, and retrieving to obtain an associated sub-graph corresponding to the subject entity; screening out a plurality of entities with the highest association degree with the subject entity from the association sub-graph, and generating a plurality of sub question sentences after complex statement decomposition according to association semantic information corresponding to each entity; according to the method, engineering generation of the decomposed question sub-sentences is realized by combining the knowledge graph with the syntactic template, the problems of poor flexibility, limited visual field and the like existing in a decomposition method purely depending on a rule template are solved, and the method has certain theoretical reference and practical reference values.

Description

Technical field [0001] The present application relates to the technical field of natural language processing, and more particularly to a complicated questioning method, apparatus and computer equipment for vertical field. Background technique [0002] With the development and application of Natural Language Generation, NLG technology, the decomposition of complex question has become an important branch task in natural language processing. The basic connotation is to break up a specific complex question in a number of subsis, which can be built into a full semantic syndrome by natural language understanding and natural language generation technology. Its main significance is: on the one hand, it will be blurred, and the rough complex question is decomposed into a number of simple sub-question. Compared with the answer, it is easier to implement the correct handling of the complex question solution; on the other hand, the complex question is broken down into a simple sub-question o...

AI Technical Summary

Problems solved by technology

However, with the increase of information elements in complex questions, the inexplicability of complex question decomposition methods based on machine learning is becoming more and more obvious, resulting in poor decomposition effects and the reality of the "black box effect".

Images