Medical Big Data Mining And Analysis Based On Doctor Cooperation

Due to the rapid development of the Internet,big data analysis,artificial intelligence and other technologies,the amount of data information in all walks of life has shown a linear growth trend,and medical resources,as a key member of the massive Internet information,have also experienced an "explosive" growth rate,making It is difficult for users and users to quickly find the guidance and guidance information they need in the massive medical information,which is also the biggest disadvantage shown in the network medicine of "information explosion" and "information overload".Based on the problems found above,this research focuses on doctors,drugs,and patients,with big data analysis-information consultation-knowledge management as the main line,starting from a series of questions such as doctor’s advice,medical assistance,and patient question-and-answer.He developed theories of big data mining,machine learning and artificial intelligence,constructed a network diagram of doctor cooperation relationship,and proposed a clustering-based doctor cooperation relationship community discovery algorithm,established a knowledge map based on disease information,and designed a medical knowledge map based on medical knowledge.The intelligent question and answer,by exploring the relationship within the doctor’s cooperation,can better serve and guide the doctor’s cooperation within the doctor,assist the doctor to deal with the patient’s medical problems more efficiently and accurately,and use the knowledge map technology to efficiently organize massive medical information.,to establish an interactive intelligent medical question answering system to assist users in completing preliminary self-diagnosis.The goal is to improve the quality of medical care services,rationally allocate and utilize medical network resources,relieve the pressure of medical staff treatment to a certain extent,and further improve medical services.with quality.Therefore,the research focus of this thesis is to use algorithms such as big data mining and machine learning to deeply explore medical information such as diseases,medications,patients and experts.Taking doctors as the research object,firstly establish a doctor’s cooperative relationship network,and through clustering community discovery algorithms The complex network of doctor cooperative relationships formed is divided into communities,and then a medical group is constructed for doctors in the same set,and doctor experts are recommended according to the communities they divide.The second is to build a knowledge map through disease data,and finally an interactive intelligent medical question answering system is established to assist patients in initial self-diagnosis.This thesis first obtains medical information from multi-source medical information platforms through web crawler technology,and verifies,fuses and organizes the medical information content obtained from different medical platforms.Physician Partnership Network.A parallelized Louvain algorithm is proposed.Next,the strategy of selecting the most suitable community discovery algorithm from four clustering methods,such as parallel Louvain algorithm,Louvain algorithm,GN algorithm and spectral clustering,divides the original network into several sub-communities.,the correlation between doctors within the sub-community is higher.Use Gephi visualization software to visualize the doctor’s partnership network,and mark the doctor with the highest median value in the sub-community as the recommended doctor needed by the patient.The second step is to perform knowledge extraction,knowledge fusion and knowledge storage on the obtained medical data to construct a medical knowledge graph.Finally,for the question of natural language composition proposed by the user,the Aho-Corasick automaton algorithm is used to extract the entities in the user question,and the feature word-based classification method is used to classify the intent of the user question,and the entity and intent are parsed into Cypher sentences.Search for knowledge in the medical knowledge graph and feed back answers to patients.