Intelligent Guidance System Of Traditional Chinese Medicine Based On Knowledge Graph

In the modern information age,exploring better ways to extract and summarize professional knowledge has become a research hotspot.In the medical field,the selection of doctors has always been a typical problem in the medical guidance system.In the current medical treatment scene,patients often register with hundreds of similar doctor information in the treatment field of the system,and can not accurately select the best match for their symptoms and diseases.The problems of wrong registration and difficult registration happen from time to time.In order to solve these common problems,this paper analyzes the needs of a specific municipal hospital as background proceeding from the actual situation,and combines the data from outpatient medical record and the data crawled from authoritative Internet websites,finally studies and develops a set of TCM intelligent guidance system based on knowledge graph.In the real-life doctor-patient question-and-answer scene,the symptom entity text that can be used for analysis in the patient’s symptom description is usually contained in a large number of meaningless natural languages.Most of the descriptions have the problems of random content and more colloquial words,which creates difficulty in the process of intelligent diagnosis.In this study,from the perspective of how to get the entity information of symptoms in the text of patients’ natural language description,an intelligent guidance system solution is proposed,which can make the user and the system have less interaction rounds.At the same time,in the actual scene of patients,there is still a problem:the doctor’s diagnosis of patients is usually not unique,and the clinical symptoms of patients are usually not unique.The manifestations and difficulties of this problem are mainly reflected in the real medical records of TCM outpatient medical records obtained in this paper.This data mainly records the many to many relationship between disease entity and symptom entity.In addition,there are problems such as large amount of data,non-uniform data form,and many dirty characters.In order to effectively apply this batch of valuable data,this paper first cleans and segments the data,and then proposes a set of data oriented algorithm to extract one to many relationship from many to many relationship,finds out the clinical symptoms and signs of all diseases covered by the data set,and establishes the knowledge map.The intelligent guidance system constructed in this paper uses named entity recognition technology around its business characteristics,and combines with knowledge mapping technology to complete the overall business logic.Focusing on these two main tasks,this paper introduces the relevant technical knowledge,and experimentally analyzes the feasibility of applying the relevant technology in this scenario.After understanding the main difficulties of the entire process,this study proposes a complete business process implementation plan and completes it in practice.Experiments have proved that the TCM intelligent diagnosis guidance system proposed in this paper can help the diagnosis function of patient’s symptom self-report well-realized and direct patients to the corresponding doctors and departments.