Multi-Source Data-Driven Personalized Recommendation and Decision-Making for Automobile Products Based on Basic Uncertain Information Order Weighted Average Operator DOI Open Access
Yi Yang, Mengqi Jie,

Jiajie Pan

et al.

Sustainability, Journal Year: 2025, Volume and Issue: 17(9), P. 4078 - 4078

Published: April 30, 2025

The extensive electronic word-of-mouth (eWOM) data generated by consumers encapsulates authentic product experience information. By leveraging advanced analysis technologies, enterprises can extract sustainable consumer behavior preference knowledge, thereby supporting the optimization of their marketing and management strategies. However, existing data-driven ranking processes predominantly focus on single-source eWOM rarely mine insights from a multi-source perspective. Moreover, quality cannot be overlooked. Consequently, this study uses automobile products as case example integrates rating data, complaint safety test to construct personalized recommendation algorithm. Specifically, an evaluation index system is established for each three types. To model information quality, these are transformed into basic uncertain (BUI), which incorporates scoring credibility metrics. XLNet employed convert text targeted models developed assess reliability Subsequently, BUI aggregated using ordered weighted average (BUIOWA) aggregation operator. Based this, method aligned with user preferences proposed, offering results that match preferences. Finally, illustrative example, elucidates process provides managerial implications enterprises.

Language: Английский

Multi-Source Data-Driven Personalized Recommendation and Decision-Making for Automobile Products Based on Basic Uncertain Information Order Weighted Average Operator DOI Open Access
Yi Yang, Mengqi Jie,

Jiajie Pan

et al.

Sustainability, Journal Year: 2025, Volume and Issue: 17(9), P. 4078 - 4078

Published: April 30, 2025

The extensive electronic word-of-mouth (eWOM) data generated by consumers encapsulates authentic product experience information. By leveraging advanced analysis technologies, enterprises can extract sustainable consumer behavior preference knowledge, thereby supporting the optimization of their marketing and management strategies. However, existing data-driven ranking processes predominantly focus on single-source eWOM rarely mine insights from a multi-source perspective. Moreover, quality cannot be overlooked. Consequently, this study uses automobile products as case example integrates rating data, complaint safety test to construct personalized recommendation algorithm. Specifically, an evaluation index system is established for each three types. To model information quality, these are transformed into basic uncertain (BUI), which incorporates scoring credibility metrics. XLNet employed convert text targeted models developed assess reliability Subsequently, BUI aggregated using ordered weighted average (BUIOWA) aggregation operator. Based this, method aligned with user preferences proposed, offering results that match preferences. Finally, illustrative example, elucidates process provides managerial implications enterprises.

Language: Английский

Citations

0