Question-answering method and system based on multi-model fusion

Abstract

The embodiment of the invention discloses a question-answering method and system based on multi-model fusion. The method comprises the following steps: constructing a knowledge base and a knowledge graph, conducting question analysis on input original questions in combination with the knowledge base and the knowledge graph, obtaining question analysis data, retrieving the question analysis data based on a matching method indicated by a fusion model, and obtaining question answers corresponding to the original questions. According to the method, higher efficiency can be achieved on the premisethat accuracy is guaranteed in the process of searching for the answers corresponding to the original questions through the fusion model, the deep learning model is increasingly robust along with increase of the data volume, the overall model understands semantics more accurately, and real intelligence can be injected to the robot.

Description

technical field

[0001] The invention relates to the technical field of artificial intelligence question answering systems, in particular to a question answering method and system based on multi-model fusion. Background technique

[0002] With the development of big data and deep learning technology, it will no longer be a fantasy to create an automatic human-machine dialogue system as our personal assistant or chat partner.

[0003] At present, dialogue systems are attracting more and more attention in various fields, and the continuous progress of deep learning technology has greatly promoted the development of dialogue systems. For dialogue systems, deep learning techniques can utilize large amounts of data to learn feature representation and response generation strategies, which only require a small amount of manual operations. Existing robots based on intent recognition and dialogue management require a large amount of labeled corpus for feature learning. The labeling p...

Images