Online reviews have emerged as a pivotal data source for product evaluation and selection. However, converting their inherent textual uncertainty into reliable information for decision support remains a significant challenge. This paper proposes a rough‐cloud‐integrated multiattribute three‐way decision (MATWD) framework that captures the uncertainty of online reviews and supports product classification and ranking. First, a rough‐cloud model is developed to transform sentiment orientations from online reviews into normal cloud representations, capturing both intrapersonal uncertainty and interpersonal uncertainty inherent in user‐generated content. Second, we introduce symmetric Kullback–Leibler divergence and probability of superiority as two novel measurements for cloud models, which enable the derivation of relative profit functions, risk avoidance coefficients, and conditional probabilities within a cloud‐based three‐way decision framework. Third, a cloud‐based indifference threshold‐based attribute ratio analysis (ITARA) method is developed to determine objective attribute weights, which are then integrated into the MATWD framework to obtain three‐way classifications and, subsequently, a comprehensive ranking of alternatives via the Borda count method. Finally, the proposed methodology is validated through a case study on new energy vehicle selection, demonstrating its practical applicability and effectiveness.
Jia et al. (Thu,) studied this question.