With the growth of the Internet and artificial intelligence,the design of intelligent expert systems to achieve automatic online consultation has become a development direction of Internet intelligent remote hospital.The goal of this thesis is to design a medical automated question and answer system based on knowledge mapping to help people more effectively access medical health knowledge with orthopedic rehabilitation as the main focus.We use Neo4j to build the graph database due to its easy-to-use,fast,visualization and other features.We collect data through three methods:collecting semi-structured data via the Internet,summarizing books,and consulting doctors.The Internet collection method is fast and large data volume,which is mainly taken in this way.Scrapy web crawler software was used to capture various medical information on the network,especially for orthopedic rehabilitation treatment knowledge,and these medical data were integrated into disease-centric structured knowledge through data cleansing and stored in the graph database Neo4j,which has seven categories of about 96,000 entity nodes,ten categories of about 1.09 million relational magnitudes and nine types of attributes.The question and answer system is implemented in the form of Chinese participle,entity recognition,semantic recognition,keyword matching,etc.At the same time,to reduce the impact of misspelled words on the answer query,a two-way maximum match algorithm is borrowed from the Chinese participle to perform fuzzy queries.Orthopaedic rehabilitation is an integral part of the overall treatment plan for all types of orthopaedic diseases and is important for preoperative preparation and postoperative recovery,for which there is very little research.Therefore,in the construction of the knowledge map,we focus on adding systematic knowledge of orthopaedic rehabilitation treatment,forming a question and answer system with general medical knowledge as the background and orthopaedic rehabilitation treatment disease knowledge as the main content,which can be professionally queried in depth,laying the foundation for Internet remote treatment in orthopaedic rehabilitation. |