Building Adaptive Linguistic Consensus for Product Selection Using Coupled Multi-Platform Online Reviews and Ratings

The rapid advancement of networks has facilitated the widespread dissemination of online reviews derived from user-generated content. These information-dense reviews have emerged as a critical decision-support resource in various domains such as product selection and recommendation. Although prior studies have integrated numerical ratings with textual reviews, effective methods for aggregating these data in a multi-platform consensus analysis remain underexplored, especially in multi-platform settings where reviews and ratings may be scattered and inconsistent. This oversight not only results in an incomplete utilization of the available decision-relevant information but may also introduce bias and reduce the reliability of decision outcomes. To address this gap, this paper proposes a novel consensus-driven group decision-making framework that integrates online reviews and ratings collected from multiple platforms. The proposed framework combines textual reviews, numerical ratings and information credibility by employing an improved representation of basic uncertain linguistic information, aggregates review information at the platform level, and treats each platform as an individual decision-maker. A consensus model is then used to reduce inconsistencies among platform-level evaluations and obtain an acceptable collective product ranking. Finally, a product-selection case study based on real-world online data validates the applicability and effectiveness of the proposed framework through comprehensive comparative analyses. The results indicate that incorporating coupled multi-platform online reviews and ratings through a consensus-oriented analytical framework yields more comprehensive and transparent decision support compared to reliance on a single information source.