The Design And Implementation Of Public Praise Analysis System For Doctor Based On Network Data

With the rapid development of Internet technology,various industries are beginning to integrate with the Internet,of course,the medical industry is no exception.The online medical website has accumulated a large number of patients' evaluation information,which is the basis for the patient to choose a doctor.The patient generally can decide which doctor to choose after reviewing the evaluation of other patients.But it often takes a lot of time and effort for the patient and it is impossible for the patient to read through all the doctor's comments.Based on this problem,a doctor's word-of-mouth analysis system based on network data is proposed to provide patients with reference indexes for selecting doctors and help patients to quickly select doctors.The system collects the patient's comment information from the internet and then conducts a text analysis.Finally,the doctor's word-of-mouth score,the proportion of satisfactory evaluation,the proportion of general evaluation and the proportion of unsatisfactory evaluation are obtained.The system is divided into information module,comment text preprocessing module,comment text module,information display module and other modules.The data crawler module uses an open source crawler framework,Web Magic,to gather information from hospitals,doctors,patient reviews,and more on medical websites.The comment text preprocessing module is responsible for commenting on text segmentation and text-to-vector conversion.The word segmentation tool uses the Ansj segmentation tool.The text mining module reviews the classification of algorithmic comments through logistic regression.Comments will eventually be divided into "satisfaction","general","not satisfied" three categories.The category of comment serves as the basis for satisfaction ratings.The information display module is responsible for displaying information such as doctor information crawled,comment information and reputation score.Display module using B / S architecture,the use of Spring MVC,Spring,My Batis and other mainstream Web development framework to build a Web server.The entire system uses the My SQL database for data storage.In the test environment,the system collected more than 1000 doctors' data and more than 50,000 comment data from the Guahao network,and trained a review classification model.The accuracy of the classification model is about 86%,which meets the needs of system application.Then The classification of evaluation and the calculation of the doctor's word of mouth score is achieved.Lastly,the display function of word-of-mouth analysis results is realized.The user can log in to the system to see the doctor's information and the doctor's score.The whole system achieved the expected function.