Construction And Application Of Knowledge Map In Internal Medicine

With the rapid development of society,the popularization of the Internet and the progress of human civilization,people have more and more diverse ways to acquire knowledge.Contemporary fast-paced life make human health problems,more and more attention to the medical science knowledge of the Internet search demand grew rapidly,but in terms of current situation,the existing medical knowledge on the Internet is rather complex,it’s very difficult for the average user from vast amounts of data in precise positioning to the required content,and the traditional way of access to information is to use a search engine,A large amount of web page data information is returned through keywords,and there will be some irrelevant information,which requires a lot of time and energy to find the answer from the returned results.Combined with the national strategic planning and layout in the medical field,it is a very correct choice to construct the demand for smart medical treatment.Therefore,it is a very correct choice to apply the knowledge graph technology to the development of the medical industry in today’s society.Therefore,this paper studies the knowledge graph technology,constructs the internal medicine knowledge graph,and the constructed knowledge graph is applied to the knowledge question answering system,so as to provide users with more efficient and convenient access to internal medicine knowledge.The main work content of this paper includes the following aspects.First of all,the collection and processing of data.The data set for this paper is made up of two parts: semi-structured data and structured data.The semi-structured data is used to capture the data in the medical website by using crawler-related technology and transform it into structured data through data processing.This part of data is used for the construction of the knowledge map of internal medicine.Structured data is acquired from foreign websites,which is used to construct the question-answering module of knowledge question-answering system.Secondly,based on the construction of internal medicine knowledge map.The Word2 Vc model was replaced by Bbert pre-training model as word vector input,and some of the deep learning models of BILSTM,CRF,BIGRU and ATTENTION were used for model fusion to carry out knowledge extraction,and then the knowledge was fused and stored in the Neo4 j graph database.To complete the construction of the knowledge map of internal medicine.In the end,build a question-answering system based on the knowledge of internal medicine.The knowledge map of internal medicine is used as the knowledge base of question answering system,through the classification of users’ questions,the user intention was studied by using the BERT-Bi GRU-Attention fusion model,and the construction of the question answering system was completed.This paper successfully completed the construction of the internal medicine knowledge map,and applied to the internal medicine knowledge question answering system,the experimental results prove that the internal medicine knowledge question answering system can meet the user’s acquisition of internal medicine knowledge.