Design And Implementation Of Medical Question Answering System Based On Knowledge Graph

As people pay more attention to health,they are more inclined to search health information on the Internet.But,when people search for related information,the search engines will return too many results and do not really understand the user's intentions,and medical websites are too professional and complicated.Ordinary users need more time and effort to find the information they need,and they can't get the information they really care about quickly and accurately.Therefore,in order to alleviate the problems above,we design and implement a medical question and answering system based on knowledge graph.Firstly,we construct a medical knowledge graph.We acquire knowledge from medical websites,and using rules and dictionary to extract medical entity and their relationship.We organize and store them in Neo4j database,it contains a variety of entities and relationships related to disease.Secondly,we design and implement the question and answering system algorithm based on the medical knowledge graph above.We use entity recognition and attribute linking algorithm based on Long Short-Term Memory(LSTM)network to calculate problem intent and related relationship attributes.The experimental results show that the accuracy rate reaches 87.3%.Combining with the medical knowledge graph search,it can be applied to the Q&A module process.Finally,based on the medical knowledge graph and related algorithms constructed above,we develop and implement a medical question and answering system based on knowledge graph,including user medical question and answer service and medical knowledge graph visualization application module.At present,the instant medical question and answering system for the public is very rare.So the system constructed in this paper can provide convenient medical question and answer service for the public to alleviate the above problems,and it has important application value.Secondly,the medical health knowledge graph in the Chinese field is rare today.The medical knowledge graph constructed in this paper can be used as the knowledge base of applications such as intelligent question answering or search,and it is convenient to expansion and migration and has a wide range of application prospects.