Intelligent question-answering method and device based on pharmaceutical knowledge graph

Abstract

The embodiment of the invention provides an intelligent question-answering method and device based on a pharmaceutical knowledge graph. The method comprises the steps of obtaining an answer to a current question; if it is judged that the answer to the current question is matched with the information represented by the current to-be-answered node, outputting the question according to the next to-be-answered node, wherein the current question is output according to the current to-be-answered node, the to-be-answered node is an intermediate node of which the expressed information cannot be acquired according to the initial input information, the intermediate node is an intermediate node on a regular path, the rule path is obtained by searching the pharmaceutical knowledge graph according to the initial input information, and the information represented by the starting node on the rule path is matched with the disease information of the patient in the initial input information. According to the intelligent question-answering method and device based on the pharmaceutical knowledge graph, intelligent questioning and answering are conducted through the pharmaceutical knowledge graph, thequestioning and answering process conforms to professional medication rules, and the accuracy and practicability of intelligent question-answering can be improved.

Description

technical field [0001] The present invention relates to the field of computer technology, and more specifically, to an intelligent question answering method and device based on a pharmaceutical knowledge graph. Background technique [0002] Intelligent question answering device has gradually become a new trend in realizing the mutual communication between human and machine, and it is an information service system that includes natural language processing, information retrieval, data mining and other interdisciplinary fields. The intelligent question answering device firstly understands the questions raised by the users, then searches the relevant data in the Internet, and finally answers the questions of the users. The working principle of the traditional question-answering device is: use crawler technology to crawl relevant question-and-answer sentences from various search engines, related forums or web pages, and integrate them into text information as a corpus. When the ...

AI Technical Summary

Problems solved by technology

[0005] Embodiments of the present invention provide an intelligent question answering method and device based on pharmaceutical knowledge graphs, to solve or at least partly solve the defect of low question answer accuracy existing in the prior art

Images