UNICON-LSGDM: A UNIform consensus framework with community detection and weighted similarity for large-scale group decision-making.
Saved in:
| Title: | UNICON-LSGDM: A UNIform consensus framework with community detection and weighted similarity for large-scale group decision-making. |
|---|---|
| Authors: | Dutta, Sangita1 (AUTHOR) 2020cspr030.sangita@students.iiests.ac.in, Ghosh, Pratyay2 (AUTHOR) pratyay.ghosh2022@uem.edu.in, Kule, Malay1 (AUTHOR) malay.kule@gmail.com, Chakraborty, Susanta1 (AUTHOR) sc@cs.iiests.ac.in |
| Source: | Sādhanā: Academy Proceedings in Engineering Sciences. Jun2026, Vol. 51 Issue 2, p1-11. 11p. |
| Subjects: | Group decision making, Consensus (Social sciences), Stated preference methods, Social network analysis, Cluster analysis (Statistics) |
| Abstract: | Large-scale group decision-making (LSGDM) is essential for resolving complex problems involving numerous stakeholders with diverse preferences, especially in domains such as governance, urban planning, and policy development. However, achieving efficient consensus in LSGDM remains challenging due to issues like preference heterogeneity, noncooperative behaviours, and the computational burden of aggregating large-scale inputs. Traditional clustering and penalty-based methods often neglect social dynamics or compromise preference authenticity. This study proposes a novel LSGDM framework that integrates a modified cosine similarity measure weighted by alternative importance, the Louvain community detection algorithm based on social ties, and a uninorm-based weight adjustment mechanism to manage cooperation levels. By leveraging social network structures to form cohesive subgroups and preserving preference diversity through priority-weighted aggregation, the framework improves consensus efficiency and stability. Experiments on networks with a varied range of decision- makers demonstrate superior performance, achieving consensus levels above 0.91 in just four iterations, outper- forming baseline methods in both speed and stability. This integrated approach addresses key limitations in existing models and offers a scalable, adaptive solution for real-world LSGDM scenarios. [ABSTRACT FROM AUTHOR] |
| Copyright of Sādhanā: Academy Proceedings in Engineering Sciences is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Large-scale group decision-making (LSGDM) is essential for resolving complex problems involving numerous stakeholders with diverse preferences, especially in domains such as governance, urban planning, and policy development. However, achieving efficient consensus in LSGDM remains challenging due to issues like preference heterogeneity, noncooperative behaviours, and the computational burden of aggregating large-scale inputs. Traditional clustering and penalty-based methods often neglect social dynamics or compromise preference authenticity. This study proposes a novel LSGDM framework that integrates a modified cosine similarity measure weighted by alternative importance, the Louvain community detection algorithm based on social ties, and a uninorm-based weight adjustment mechanism to manage cooperation levels. By leveraging social network structures to form cohesive subgroups and preserving preference diversity through priority-weighted aggregation, the framework improves consensus efficiency and stability. Experiments on networks with a varied range of decision- makers demonstrate superior performance, achieving consensus levels above 0.91 in just four iterations, outper- forming baseline methods in both speed and stability. This integrated approach addresses key limitations in existing models and offers a scalable, adaptive solution for real-world LSGDM scenarios. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 02562499 |
| DOI: | 10.1007/s12046-026-03103-x |
