Abstract
Online medical platform is a platform for patients to post their medical experience, collect medical in-formation, and link doctors and patients for related medical activities. As the number of patients and doctors registered on the platform increases, there is an urgent need to consider how patients can iden-tify useful information from the vast amount of information to help them make medical choices, and how the platform can provide distinctive medical choices based on the risk preferences of patients. In this paper, we propose a decision-making model that integrates a machine-learning method and a multi -criteria decision-making method to help patients to select physicians based on user-generated content considering interactive criteria and risk preferences of patients. Firstly, by data mining methods, vari-ous criteria included in user-generated content that influence patients' selection behavior are retrieved to construct a physician evaluation system. Secondly, a sentiment analysis method based on a medical review corpus is developed to mine satisfaction information from text reviews. Finally, a multi-criteria decision-making method is proposed considering patients' risk-averse preferences for medical services and the interactions among criteria. The validity of the proposed model is demonstrated by a case study of ranking psychologists from haodf.com. The comparisons with other methods and sensitivity analysis results provide suggestions to patients to choose psychologists and the website to rank psychologists.(c) 2022 Elsevier Ltd. All rights reserved.