Intelligent question answering method and device for multi-source heterogeneous data source

Abstract

The invention provides an intelligent question answering method and device for a multi-source heterogeneous data source. The method comprises the following steps of: calculating a first similarity between an input question and a candidate relationship; obtaining a first answer from the knowledge base according to the first similarity; retrieving a multi-hop text according to the input question; calculating a second similarity between sentences in the multi-hop text and the input question and a third similarity between entities in the multi-hop text and the input question; obtaining a second answer from the multi-hop text according to the second similarity and the third similarity; and determining a relationship between the first answer and the second answer, and outputting an answer text of the input question. According to the intelligent question answering method and device for the multi-source heterogeneous data source, the accuracy and richness of the output answers are improved, and a simple and complete answer which is easier to understand by a user can be returned.

Description

technical field [0001] The present invention relates to the technical field of deep learning and intelligent question answering, in particular to an intelligent question answering method and device for multi-source heterogeneous data sources. Background technique [0002] The realization of intelligent question answering involves multiple technologies such as knowledge base question answering, information retrieval, and knowledge representation, and has been gradually applied to scenarios such as knowledge acquisition and chat robots. According to the source of question answers, question answering technology can be divided into two directions, one is knowledge base question answering technology whose answers come from structured information; the other is text question answering technology whose answers come from unstructured information. [0003] On the one hand, with the rapid development of deep learning technology, more and more researchers are trying to use deep neural n...

AI Technical Summary

Problems solved by technology

However, these methods only focus on the entity information in the text; lack of filtering of noise sentences, etc., thus reducing the reasoning ability of the model

[0005] In addition, the existing intelligent question answering technology rarely studies the method of using multi-source heterogeneous information sources to combine the two to complement each other.

Images