Research And Implementation Of Disease Knowledge Question Answering System Based On Knowledge Graph

The 21 st century is a highly information-based era.The rapid development of the mobile Internet has had a significant impact on all areas of our lives.How to acquire information correctly and efficiently becomes an important prerequisite for today’s intelligent information age.Knowledge graph provides a means to extract structured knowledge from massive data.Nowadays,with the rapid development of knowledge graph technology,it has been among the hot research directions of natural language processing and has been widely used in many fields such as intelligent search and intelligent question answering.With the improvement of the quality of life,the demand for Internet medical science search has increased dramatically.As an inevitable product of the development of artificial intelligence in the information age,question answering systems can directly understand the user’s intentions and can make targeted answers based on different input from users,providing users with great convenience.Therefore,this paper implements a disease knowledge question answering system based on knowledge graph.The specific research content of this article mainly includes the following two aspects:Study how to build a knowledge graph of diseases in the medical field,which mainly includes knowledge extraction and knowledge storage.Firstly,use the crawler tool to crawl disease knowledge data from a healthy website and organize the captured data into json format.Then,design entities in the medical field and their relationships based on the crawled disease information.Finally,build a comprehensive disease knowledge graph in the medical field with the help of the graph database Neo4 j.Study how to construct a disease knowledge question answering system in the medical field,which mainly includes question analysis and answer query.Question analysis is mainly responsible for dividing the user’s question into the corresponding question type.The answer query is responsible for converting the question sentence into the corresponding Cypher query sentence according to the designed query template to query the answer from the knowledge graph.For the questions that need to be answered,this paper chooses to use the AC automaton to complete the medical entity recognition task,use the rule base to match questions and calculate the similarity of word vectors and strings to complete the question classification task.