This book explores clustering operations in the context of social networks and consensus-reaching paths that take into account non-cooperative behaviors. This book focuses on the two key issues in large-scale group decision-making: clustering and consensus building. Clustering aims to reduce the dimension of a large group. Consensus reaching requires that the divergent individual opinions of the decision makers converge to the group opinion.
This book emphasizes the similarity of opinions and social relationships as important measurement attributes of clustering, which makes it different from traditional clustering methods with single attribute to divide the original large group without requiring a combination of the above two attributes. The proposed consensus models focus on the treatment of non-cooperative behaviors in the consensus-reaching process and explores the influence of trust loss on the consensus-reaching process.The logic behind is as follows: firstly, a clustering algorithm is adopted to reduce the dimension of decision-makers, and then, based on the clusters’ opinions obtained, a consensus-reaching process is carried out to obtain a decision result acceptable to the majority of decision-makers.
Graduates and researchers in the fields of management science, computer science, information management, engineering technology, etc., who are interested in large-scale group decision-making and consensus building are potential audience of this book. It helps readers to have a deeper and more comprehensive understanding of clustering analysis and consensus building in large-scale group decision-making.
Author(s): Su-Min Yu, Zhi-Jiao Du
Publisher: Springer
Year: 2022
Language: English
Pages: 203
City: Singapore
Preface
Contents
About the Authors
Acronyms
Symbols
List of Figures
List of Tables
1 Introduction
1.1 Motivation
1.2 Chapter Overview
2 Preliminary Knowledge
2.1 Large-Scale Group Decision-Making (LSGDM) and Clustering
2.2 Social Network Analysis (SNA)
2.3 Consensus Building and Non-cooperative Behaviors
2.3.1 Consensus Building
2.3.2 Non-cooperative Behaviors
3 Trust-Similarity Analysis-Based Clustering Method
3.1 Introduction
3.2 TSA-Based Decision Information Processing
3.2.1 Establishment of the Trust-Similarity Matrix
3.2.2 Construction of the TSA Plot
3.2.3 Determination of Crosshair Placement
3.3 TSA-Based Clustering Algorithm
3.3.1 Definition of the Joint Threshold
3.3.2 Algorithm Design of the TSA-Based Clustering Method
3.3.3 Time Complexity Analyses
3.4 Numerical Experiment
3.5 Further Analysis of the TSA-Based Clustering Method
3.5.1 Calculation of the Joint Threshold
3.5.2 Simulation with Different Methods for Calculating the Crosshair Placement
3.5.3 Simulation with Different Measurement Attributes
3.6 Comparison with Other Clustering Methods
3.6.1 Comparison with the K-Means Clustering Method and Xu and Chen's Method XuspsChenspsSEEsps2005
3.6.2 Comparison with Dong et al.'s Method DongspsDingspsINSsps2017 and Ding et al.'s Method DingspsChenspsINSsps2019
3.7 Conclusions
4 Trust-Similarity Measure-Based Hierarchical Clustering Method
4.1 Introduction
4.2 Problem Configuration and Decision Information Preprocessing
4.3 Agglomerate Hierarchical Clustering Method with Given Constraints
4.3.1 Only the Constraint of the Trust-Similarity Score (Type A)
4.3.2 Trust Priority or Similarity Priority Constraint (Type B or Type C)
4.3.3 Constraint with No Bias Between Trust Relationship and Opinion Similarity (Type D)
4.4 Visualization of Clustering Process
4.5 Generation of the Weights and Opinions Regarding Clusters
4.6 Numerical Experiment
4.7 Comparative Analysis and Discussions
4.7.1 Analysis of the Determination of Clustering Constraints
4.7.2 Comparison of Clustering Results with Different Types of Constrains
4.7.3 Comparison with Other Clustering Methods
4.7.4 In-Depth Analysis of the Calculation of Weights Regarding Clusters
4.8 Conclusions
5 Hierarchical Punishment-Driven Consensus Model for Probabilistic Linguistic LSGDM
5.1 Introduction
5.2 Problem Framework Configuration and Preprocessing of Decision Information
5.3 Hierarchical Clustering Algorithm
5.4 Hierarchical Punishment-Driven Consensus-Reaching Model in PL-LSGDM Problems
5.4.1 Consensus Measure
5.4.2 Punishment-Driven Consensus Iterations
5.4.3 Hierarchical Punishment-Driven Consensus Iterations
5.5 Case Study
5.5.1 Problem Description
5.5.2 Decision Process
5.6 Comparative Analysis and Discussion
5.6.1 Further Discussion of the Hierarchical Punishment-Driven Consensus Model
5.6.2 Comparison to Other Linguistic Models
5.6.3 Managerial Implications
5.7 Conclusions
6 Confidence Consensus-Based Model for LSGDM
6.1 Introduction
6.2 Consensus Measures
6.3 Mechanism for Managing Non-cooperative Behaviors
6.3.1 Identification Rule for Non-cooperative Behaviors
6.3.2 Interaction and Discussion
6.3.3 Proper Modification
6.3.4 Algorithm for the Confidence Consensus-Based Model in LSGDM
6.4 Case Study
6.5 Comparative Analysis
6.5.1 Effect of the Confidence Consensus-Based Model on Managing Non-cooperative Behaviors
6.5.2 Analysis of Different Consensus Thresholds
6.5.3 Comparison with Xu et al.'s Model XuspsDuspsDSSsps2015 and Zhang et al.'s Model ZhangspsDongspsTFSsps2017
6.5.4 Comparison with Different Consensus Measures
6.5.5 Comparison with Different Parameters
6.6 Conclusions
7 Integration of Independent and Supervised Consensus Models
7.1 Introduction
7.2 Analysis of Consensus Level in LSGDM
7.2.1 Opinion Clustering
7.2.2 Consensus Measure
7.3 Proposed Consensus-Reaching Models for Managing Non-cooperative Behaviors
7.3.1 Independent Consensus-Reaching Model
7.3.2 Supervised Consensus-Reaching Model
7.3.3 Mixed Consensus-Reaching Model
7.4 Case Study
7.5 Comparative Analysis and Discussion
7.5.1 Sensitivity Analysis
7.5.2 Comparative Analysis of ICRM, SCRM, and MCRM
7.5.3 Comparative Analysis of Four Punishment Approaches to Non-cooperative Behaviors
7.5.4 Comparison with Xu et al.' Model XuspsDuspsDSSsps2015
7.6 Conclusions
8 Consensus Building: Coordination Between Trust Relationships and Opinion Similarity
8.1 Introduction
8.2 Consensus Measure
8.3 Consensus-Reaching Process Considering Trust Loss
8.3.1 Identification of Non-cooperative Behaviors
8.3.2 Management of Non-cooperative Behaviors
8.3.3 Decision-Making Algorithm Based on Two-Dimensional Consensus Model
8.4 Numerical Experiment
8.5 Further Discussion of Two-Dimensional Consensus Model
8.6 Conclusions
9 Conclusions and Future Research Directions
9.1 Conclusions
9.2 Future Research Directions
Appendix References
