Intelligent question-answering system and method, and related equipment

Abstract

The invention discloses an intelligent question-answering system and method, and related equipment. The system comprises: a problem analysis subsystem, which is used for generating a problem description text and identifying a problem type, i.e., a fact type, a non-type and a definition type; a knowledge base generation subsystem is used for arranging factual and non-factual knowledge data into anentity quadruple to form a knowledge bar, arranging the defined knowledge data into a question answer pairing form, and forming a knowledge bar in combination with the text feature vector; and an answer extraction and generation subsystem used for analyzing the fact type and non-type questions to obtain an entity quadruple, performing retrieval matching to obtain an answer, and performing vectorization processing and feature vector similarity calculation on the defined question to obtain an answer. Question classification is carried out, database query and feature vector similarity calculationare respectively used for matching, and questions are answered more accurately; and the knowledge content library adopts a structured knowledge strip form, so that the storage amount can be reduced,and the retrieval efficiency is higher and more convenient.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to an intelligent question answering system and method and related equipment. Background technique [0002] With the development of information technology, artificial intelligence technology, and Internet technology, the technology of interaction between human beings and intelligent systems has been continuously improved, and related intelligent applications have developed rapidly. [0003] In recent years, intelligent question answering system is a research hotspot in the field of artificial intelligence, and it plays an increasingly important role in human daily life. The natural language processing domain tasks contained in the intelligent question answering system usually need to process the knowledge information contained in a large amount of corpus texts. Finding a better way to realize an intelligent question answering system based on a large amount of knowle...

AI Technical Summary

Problems solved by technology

However, such a matching method is likely to lead to the situation that the user's question is not correctly understood and the wrong answer is returned. For example, the two questions of "how is the process of applying for a bonus" and "how to apply for a bonus" should ultimately point to the same answer Yes, but only the keyword matching of fuzzy query is used to identify the question, which will lead to the wrong match of the answer or the situation that the answer cannot be found

[0011] 2. The traditional knowledge base is all constructed in the form of "question-answer" pairing. In order to match the questions raised by users as much as possible, an answer will be linked with a variety of questions in different descriptions in the knowledge base, so that Answer matching, and this knowledge base construction method will increase the database capacity, resulting in the problem of reduced retrieval efficiency

Images