Font Size: a A A

Research On Personalized Service Recommendation System And Its Application In Internet Medicine

Posted on:2020-11-13Degree:MasterType:Thesis
Country:ChinaCandidate:T T LiFull Text:PDF
GTID:2404330575965398Subject:Engineering
Abstract/Summary:PDF Full Text Request
As we all know,most of the inquiries of ordinary patients are to seek help from doctors in hospitals,which brings great inconvenience to many patients who have difficulty going out or have other reasons.With the development of Internet services,personalized medical service recommendation can provide patients with a consistent and personalized medical recommendation service.However,the existing recommendation of individualized medical services for patients is based on expensive wearable equipment or health cabins similar to telephone booths as data acquisition terminals.There are not only great problems in data acquisition,but also problems of single dimension of data in wearable equipment or health cabins for various physiological indicators of patients,and complex and diverse medical resources.It is difficult to provide reliable personalized health advice for patients.Therefore,in view of the problems described above,this paper proposes a patient-oriented medical service recommendation method,which combines patient description,geographical location,various physiological indicators,disease history and other multi-dimensional data to study the user personalized service recommendation system,so as to improve the quality of medical service.(1)Firstly,aiming at the actual needs,this paper proposes a medical service recommendation system,which uses mobile devices as user data acquisition terminals and makes use of the advantages of mobile Internet to solve the problem of user’s medical consultation.The system is based on recommendation algorithm and uses patient’s condition description to recommend suitable doctors.Firstly,the system preprocesses the disease description,including word segmentation,de-noising symbols,de-stop words,word frequency statistics and text clustering.Then,on the basis of clustering,collaborative filtering algorithm is implemented to find neighboring users.Finally,the recommendation list is generated and sent to users to realize text consultation and recommendation.(2)Secondly,aiming at the actual environment and user selection problem,an improved traditional collaborative filtering algorithm based on distance correction user similarity calculation is proposed.On the basis of traditional collaborative filtering,the Pearson correlation coefficient formula for calculating user similarity is modified and the distance factor is introduced,which is determined according to the general idea of user’s close-range priority selection on the basis of balancing medical resources.Then the actual distance is used as the distance coefficient to determine the distance factor.Finally,according to the comparative analysis of the experimental results,the effectiveness of the improved algorithm for the system platform is further determined,which can provide users with more reasonable recommendations.(3)Firstly,aiming at the actual needs,this paper proposes a medical service recommendation system,which uses mobile devices as user data acquisition terminals and makes use of the advantages of mobile Internet to solve the problem of user’s medical consultation.The system is based on recommendation algorithm and uses patient’s condition description to recommend suitable doctors.Firstly,the system preprocesses the disease description,including word segmentation,de-noising symbols,de-stop words,word frequency statistics and text clustering.Then,on the basis of clustering,collaborative filtering algorithm is implemented to find neighboring users.Finally,the recommendation list is generated and sent to users to realize text consultation and recommendation.Aiming at the shortcomings of traditional internet medical service such as single data dimension,this paper combines patient consultation description,geographical location,disease history,doctor and other multi-dimensional data,based on mobile devices,uses natural language processing,text clustering and improved collaborative filtering recommendation algorithm to process user data,and realizes user personalized medical service recommendation system.It is helpful to provide personalized,reliable and accurate Internet medical services for patients,effectively reduce the problems of traditional medical services,and bring high-quality intelligent recommendation services for users.
Keywords/Search Tags:internet medicine, text clustering, user similarity, collaborative filtering
PDF Full Text Request
Related items