Research On User Satisfaction Evaluation Of Online Health Communities Based On Text Minin

In recent years,with the rapid development of the Internet,virtual communities are popularized and gradually integrated into the daily life of consumers.With the improvement of health information literacy,consumers begin to take the initiative to acquire,publish and share health-related information.The user satisfaction of online health community can reflect the development advantages and disadvantages of the current online health community.The systematic evaluation of the user satisfaction of online health community can help the online health community improve the quality of service and rationally allocate community resources so as to better meet the needs of users.It can be seen that scientific analysis and research on the user satisfaction of online health community is of great significance to enhance community services,realize sustainable development of community and rational allocation of resources and other management decisions of online health community.In this paper,the LDA model is used to cluster the user comments in online health community,and the evaluation index system of user satisfaction in online health community is constructed.The system contains 5 criteria and 14 evaluation factors.The weight of each criterion and evaluation factor is determined by the comparison of the weight of word frequency and the analytic hierarchy process.The weight ranking of criteria layer from high to low is: doctor quality,information quality,purchasing experience,system quality and service quality.On this basis,three kinds of machine learning classification models are trained,and the parameters of each model are compared and analyzed.Finally,naive Bayes is selected as the classifier in this paper.The trained naive Bayes classification model was used to conduct topic analysis and emotion analysis on the review data.Through the established online health community user satisfaction evaluation model,the satisfaction of doctor quality,information quality,purchasing experience,system quality and service quality were calculated as: Not satisfied,very satisfied,not satisfied,somewhat satisfied and not satisfied level,at the same time,the overall satisfaction level of online health community is at the basic satisfaction level.Finally,the IPA model is used to analyze the importance and satisfaction of the evaluation factors of the online health community,and it is found that the interaction speed of doctors,service level,personal privacy security and after-sales support are in the range of high importance and low satisfaction,indicating that the online health community needs to devote more energy to these four aspects of high importance but low satisfaction.Finally,according to the results of IPA model,some suggestions are put forward for the future development of online health community.