Multistage Dynamic Online Codiagnosis Three-Sided Matching Decision-Making Approach Based on Regret Theory and Evidential Reasoning

To meet the service requirements of online codiagnosis matching in a hierarchical diagnosis and treatment environment, this paper proposes a multistage dynamic three-sided matching decision-making (TSMDM) method based on regret theory and evidential reasoning, which can consider hybrid preference information and the psychological behavior characteristics of regret avoidance among matching parties. First, on the basis of reviews on the online medical platform Dingxiang Doctor in China, the objective matching attributes of patients toward grassroots doctors are obtained via Word2Vec and SnowNLP technology. Then, a corresponding weight determination model is constructed that combines these attributes and preference information for each attribute provided by decision-makers (DMs). Second, considering the fuzziness and diversity of the preference information of the matching parties, a hybrid preference information fusion model is constructed. Furthermore, the dynamics of the matching decision-making process are considered by integrating the learning mechanism into the calculation of dynamic reference points for dynamic matching problems. Third, from the perspective of the bounded rationality of matching parties, regret theory and evidential reasoning are used to obtain the degree of satisfaction of three-sided matching parties. Then, a multistage dynamic three-sided stable matching optimization model is established to maximize the corresponding degree of satisfaction of the three sides, which can obtain the final matching pairs. Finally, an example of online codiagnosis of chronic diseases under a medical hierarchical diagnosis and treatment environment is explored, and sensitivity analysis is also performed to verify the feasibility and stability of the proposed method. The results show that the proposed method can better express the preference information of three sides and improve their satisfaction with the treatment process.