Scholarlink

<p>In the era of rapid advancements in artificial intelligence and computational intelligence, the management of uncertainty remains a critical challenge, especially in complex decision-making scenarios such as industrial air-quality management. Against this backdrop, air pollution has escalated into a critical environmental threat, undermining public health and the sustainability of economic growth. Mitigating this challenge demands robust decision-making architectures capable of harmonizing heterogeneous viewpoints and driving consensus among a multitude of stakeholders. We propose a novel two-stage adaptive consensus framework that synergistically integrates three-way clustering (TWC) with fuzzy c-means (FCM) to address large-scale group decision-making (LSGDM) challenges in industrial air-quality management, specifically targeting air-quality improvement. By treating each decision-maker’s (DM) evaluation matrix as a soft-partitioning problem, our TWC-FCM hybrid first detects overlapping subgroups and partitions DMs into core and fringe regions. In Stage 1, consensus within core regions is accelerated through mixed modification strategies that leverage high-consensus fringe opinions as guidance. Stage 2 then steers fringe-region DMs toward global consensus by dynamically adjusting their weights according to an overlap-level-aware three-way weighting scheme. Experimental results on real-world air-quality data demonstrate that the model attains a consensus level of 0.9102 in merely five iterations with an adjustment cost of 0.3686, outperforming state-of-the-art alternatives in both efficiency and practicality.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Fusing three-way clustering and fuzzy c-means clustering to large-scale group decisions with adaptive two-stage consensus

  • 期刊论文

  • 发表日期：

    2025年11月10日

查看全文
  • Xin Shao,
  • Xueling Ma,
  • Muhammet Deveci,
  • Jianming Zhan

摘要

In the era of rapid advancements in artificial intelligence and computational intelligence, the management of uncertainty remains a critical challenge, especially in complex decision-making scenarios such as industrial air-quality management. Against this backdrop, air pollution has escalated into a critical environmental threat, undermining public health and the sustainability of economic growth. Mitigating this challenge demands robust decision-making architectures capable of harmonizing heterogeneous viewpoints and driving consensus among a multitude of stakeholders. We propose a novel two-stage adaptive consensus framework that synergistically integrates three-way clustering (TWC) with fuzzy c-means (FCM) to address large-scale group decision-making (LSGDM) challenges in industrial air-quality management, specifically targeting air-quality improvement. By treating each decision-maker’s (DM) evaluation matrix as a soft-partitioning problem, our TWC-FCM hybrid first detects overlapping subgroups and partitions DMs into core and fringe regions. In Stage 1, consensus within core regions is accelerated through mixed modification strategies that leverage high-consensus fringe opinions as guidance. Stage 2 then steers fringe-region DMs toward global consensus by dynamically adjusting their weights according to an overlap-level-aware three-way weighting scheme. Experimental results on real-world air-quality data demonstrate that the model attains a consensus level of 0.9102 in merely five iterations with an adjustment cost of 0.3686, outperforming state-of-the-art alternatives in both efficiency and practicality.