文献类型：期刊文献

英文题名：A novel method for large-scale group decision-making with application to e-commerce software system evaluation

作者：Yue, Chuan[1]

机构：[1] College of Mathematics and Computer Science, Guangdong Ocean University, Guangdong, Zhangjiang, 524088, China

年份：2026

卷号：191

外文期刊名：Applied Soft Computing

收录：EI(收录号：20260319935853)、Scopus(收录号：2-s2.0-105027638131)

语种：英文

外文关键词：Application programs - Big data - Data centers - Data consistency - Data reliability - Decision making - Entropy - Fuzzy rules - Number theory

外文摘要：BACKGROUND: Large-scale group decision-making (LSGDM) in big data environments faces challenges in robust data center construction, objective expert weighting, and efficient information fusion. OBJECTIVE: This study aims to develop a novel LSGDM framework integrating a Golden Ratio-based data center and an inversion-based data quality metric to improve ranking stability and decision reliability. METHODS: A GR-based data center was introduced to replace conventional mean/median centers, alongside an inversion-number-driven quality metric for expert weighting and a scalable aggregation technique for converting crisp data into intuitionistic fuzzy matrices. The framework was validated through dynamic experiments and sensitivity analysis. RESULTS: The GR-based center outperformed mean/median centers in 95% of test scenarios. The inversion-based method achieved perfect ranking consistency (Kendall's τ=1), showing a 50% improvement over entropy-based methods (τ=2/3), and maintained 100% ranking stability under parameter variations—20 times higher than entropy-based approaches. CONCLUSION: The proposed framework offers a robust, quantitatively validated solution for LSGDM in data-intensive environments, with significant advantages in consistency and scalability. ? 2026