Intelligent conversation method, device and terminal based on machine reading comprehension

Abstract

The invention provides an intelligent dialogue method, a device and a terminal based on machine reading comprehension. The method comprises the following steps: obtaining a text corresponding to a question put forward by a user; word segmentation and vectorization being performed on the problem and the text to obtain the problem vector and the text vector; inputting a problem vector and a text vector into the attention model to obtain a first vector and a second vector, wherein the first vector is used for indicating the degree of influence of the problem on noticing any word in the text, andthe second vector is used for indicating the degree of influence of the text on generating the problem; an answer start point and an answer end point being determined in the text according to the first vector and the second vector, and a paragraph between the answer start point and the answer end point being determined as the answer of the question. The method of the invention does not need to preset a 'question-answer' pair in advance, can flexibly answer various questions of a user, overcomes the defect that the prior art needs to continuously maintain the question database, and reduces thedata update cost.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence. Specifically, the present invention relates to an intelligent dialogue method, device and terminal based on machine reading comprehension. Background technique [0002] Existing intelligent dialogue systems mainly obtain answers by retrieving questions entered by users. Natural language dialogue technology based on information retrieval needs to index large-scale dialogue data in the form of "question-answer" pairs. During online dialogue, search for questions similar to the user's input and return the corresponding answer to the user. . However, when the matching degree between the user input and the corpus in the database is low, the system cannot guarantee that the system will return semantically relevant dialogues, and the existing intelligent dialogue systems have low scalability and cannot generate replies that the corpus never has. [0003] Therefore, the existing...

AI Technical Summary

Problems solved by technology

However, when the matching degree between the user input and the corpus in the library is low, the system cannot guarantee that the system will return semantically relevant dialogues. The existing intelligent dialogue systems have low scalability and cannot generate replies that the corpus has never had.

[0003] Therefore, the existing intelligent dialogue system is limited by the amount of pre-stored data in the corpus, and a large number of questions cannot be answered, especially when the user asks a new question, the system cannot answer or answer the question.

In addition, due to frequent changes in FAQs and answers, it is labor-intensive to regularly update the data in the corpus

Images