Research On Online Medical Intelligent Matching Algorithm Based On Hypergraph Learning From Two-Sided Perspective

With the development of information technology,the online medical platform represented by haodaifu,dingxiangyisheng and weiyi is getting more and more attention.Q&A community in online medicine platform has become a major way for patients to seek medical treatment because of its professionalism and convenience.With the increasing number of patients using online medical Q&A community,the number of questions generated is also increasing.At the same time,due to the limitation of time,ability and profession,doctors have limited questions to answer.Therefore,how to make full use of the existing data to improve the doctor-patient matching has become a major research hotspot.This paper takes online medical Q&A community as the research object,from the perspective of patients and doctors,to study the problem of doctor-patient matching.The main research contents are as follows:Firstly,to solve the problem of patients' personal preference learning,this paper considers that patients have both explicit and implicit preferences,and then gives different ways to describe different forms of preferences.In view of the explicit preference,the prospect theory is introduced to transform the interval preference into the satisfaction degree.For the implicit preference,the weighted time function is improved to calculate the time weight,and then the satisfaction degree is calculated based on the time weight.When multi features are fused,POWA operator is introduced to fuse the priority relationship between features.Secondly,to solve the problem of doctors' expertise learning,this paper divides the professional ability of doctors to answer questions into two parts: doctor-expertise distribution and expertise-question distribution.In the doctor-expertise distribution,the hypergraph network is constructed by using the medical relationship,and then the hypergraph is cut by using the spectral clustering,and the membership degree of doctors under each expertise is calculated,that is,the doctor-expertise distribution.In the expertise-question distribution,this paper uses representative questions to represent expertise.The expertise-question distribution is the similarity between the questions to be matched and the representative questions.Finally,for the construction of matching model,the overall satisfaction degrees of patients and doctors are maximized as bi-objective function,and doctors' ability limit and a question are answered by multiple doctors as constraints.Considering that the patients,doctors and doctors' ability status are changing in multi-stage,a dynamic matching model is established.In the model solution,because the doctor's objective function has a higher priority,the constraint method is introduced to solve model,and finally the matching results are obtained.The model proposed in this paper not only provides a new solution for doctor-patient matching,but also improves the two-sided satisfaction degrees of patients and doctors;it also ensures the quality of doctors 'answers in the online medical Q&A community,saves patients' waiting time,and relieves current tension doctor-patient relationship.Aiming at the problem of matching decision-making between doctors and patients in the online medical Q&A community,the method proposed in this paper can provide theoretical and practical reference for relevant research.