This book investigates in detail large-scale group decision-making (LSGDM) problem, which has gradually evolved from the traditional group decision-making problem and has attracted more and more attention in the age of big data. Pursuing a holistic approach, the book establishes a fundamental framework for LSGDM with uncertain and behavioral considerations. To address the behavioral uncertainty and complexity of large groups of decision-makers, this book mainly focuses on new solutions of LSGDM problems using the interval type-2 fuzzy uncertainty theory and social network analysis techniques, including the exploration of uncertain clustering analysis, the consideration of social relationships, especially trust relationships, the construction of consensus evolution networks, etc. The book is intended for researchers and postgraduates who are interested in complex group decision-making in the new media era. Authors also investigate the similar features between LSGDM problems and group recommendations to study the applications of LSGDM methods. After reading this book, readers will have a new understanding of the LSGDM study under the real complicated context.
Author(s): Tong Wu, Xinwang Liu
Series: Uncertainty and Operations Research
Publisher: Springer
Year: 2023
Language: English
Pages: 371
City: Singapore
Preface
Contents
1 Introduction
1.1 What is Large-Scale Group Decision-Making?
1.2 Literature Review Regarding Large-Scale Group Decision-Making
References
2 Uncertain Theory and Group Decision-Making
2.1 Interval Type-2 Fuzzy Sets
2.1.1 Basic Concepts of IT2 FSs
2.1.2 Distance and Similarity of IT2 FSs
2.1.3 Aggregation Operators of IT2 FSs
2.2 Group Decision-Making
2.2.1 Fuzzy Preference Relation
2.2.2 Traditional Consensus Reaching Models
2.3 Decision Behavior Analysis
2.3.1 Prospect Theory
2.3.2 Social Network Analysis
2.3.3 Trust Networks
References
3 Interval Type-2 Fuzzy Decision-Making
3.1 Interval Type-2 Fuzzy Linguistic Variables
3.2 Interval Type-2 Fuzzy Multi-Criteria Decision-Making Methods
3.2.1 Interval Type-2 Fuzzy ANP Method
3.2.2 Interval Type-2 Fuzzy TOPSIS Method
3.3 An Interval Type-2 Fuzzy Quality Function Deployment Model
3.3.1 Introduction of Quality Function Deployment
3.3.2 Classify CRs Using Kano Model
3.3.3 Determine the Weights for CRs Considering the Development Stages of Enterprises
3.3.4 Rank the Priorities of DRs Using the Extended Linguistic TOPSIS Method
3.3.5 The Proposed QFD Model and Its Application
3.4 Dimension Reduction Analysis Based on Uncertain Preference Information
3.4.1 Interval Type-2 Fuzzy Equivalence Relation Clustering Method
3.4.2 Interval Type-2 Fuzzy Principal Component Analysis
3.5 Interval Type-2 Fuzzy Trust Computing
3.5.1 Trust Degree Represented by IT2FSs
3.5.2 Interval Type-2 Fuzzy Trust Propagation and Aggregation
3.5.3 Interval Type-2 Fuzzy Trust Evaluation Model in S-Commerce
3.5.4 Illustrative Example
References
4 Large-Scale Group Decision-Making with Interval Type-2 Fuzzy Preferences
4.1 LSGDM Using Interval Type-2 Fuzzy Equivalence Relation Clustering Analysis Model
4.1.1 Procedures of the IT2-FEC-Based LSGDM Method
4.1.2 Numerical Example
4.2 LSGDM Using Interval Type-2 Fuzzy Principal Component Analysis
4.2.1 Attributes Reduction Using the IT2-PCA Model
4.2.2 An E-Commerce Case Study for the IT2-PCA-Based LSGDM Model
4.3 A Dynamic Programming Clustering Model for Interval Type-2 Fuzzy LSGDM Problems
4.3.1 Description of the LSGDM Problems
4.3.2 New Clustering Model Based on Dynamic Programming Algorithm
4.3.3 New Weight Determination Model
4.3.4 New Ranking Method Based on the Centroids of IT2 FSs
References
5 Interval Type-2 Fuzzy Large-Scale Group Decision-Making Considering Social Relationships
5.1 Main Ideas of Uncertain LSGDM Methods Considering Social Relationships
5.2 An Interval Type-2 Fuzzy TOPSIS Model for LSGDM with Social Network Information
5.2.1 Community Detection of LSGDM
5.2.2 Weights Determination for Individual Experts
5.2.3 Partition Weights Determination
5.2.4 The Solution for LSGDM Problems with Social Relationships
5.2.5 Main Characteristics of the Proposed LSGDM Model
5.3 An LSGDM Method Combining Internal Preference Information and External Social Network Structures
5.3.1 Main Ideas of the Proposed LSGDM Method
5.3.2 Combination of the Internal and External Information
5.3.3 Community Detection with Hybrid Information
5.3.4 Main Steps of the Proposed LSGDM Method
5.3.5 Illustrative Example and Comparisons
References
6 Social Trust Relationships in Group Decision-Making
6.1 CRP with the Implicit Trust Relationships
6.1.1 Cost Consensus Models Based on the Implicit Trust Rlationships
6.1.2 Modify the Adjustment Costs Considering the Trust Behavior
6.1.3 Illustrative Example
6.1.4 Comparative Analysis
6.2 Social Network Consensus with Uninorm Interval Trust Propagation
6.2.1 Identification of Consensus Inconsistency
6.2.2 Trust-Based Personalized Feedback Mechanism
6.2.3 Personalized Adoption Coefficient Based on Minimum Cost
6.2.4 Comparison and Discussion
6.3 Social Trust-Driven Consensus Considering Assessments-Modifications Willingness
6.3.1 Social Trust Analysis
6.3.2 Social Trust Driven Consensus Model with Minimum Adjustments
6.3.3 Interactive Social Trust Driven Consensus Reaching Process
References
7 Consensus Evolution Networks in Group Decision-Making
7.1 Consensus Evolution Networks
7.1.1 Concept and Characteristics of CENs
7.1.2 Consensus Measure Based on CENs
7.1.3 Feedback Adjustment Process Based on CENs
7.2 Consensus Evolution on Multiplex Networks Considering Trust Relationships
7.2.1 Trust-Consensus Multiplex Networks
7.2.2 Evolution of the Trust-Consensus Multiplex Networks
7.2.3 Overall Consensus Level of the Evolved Trust-Consensus Multiplex Networks
7.2.4 Main Procedures of the Proposed Consensus Evolution Model
7.2.5 Illustrative Example
References
8 Large-Scale Group Decision-Making Considering Trust Behavior
8.1 An LSGDM Method Based on A Two-Stage Social Trust Partition Model
8.1.1 The Shortest Path Algorithm for Trust Propagation
8.1.2 Two-Stage Trust Network Partition Algorithm
8.1.3 Weights Determination with Trust Information
8.1.4 Main Procedures and Characteristics of the LSGDM Model
8.2 Non-cooperative Behavior Management in LSGDM with Trust Relationships
8.2.1 Non-cooperative Behaviors Analysis and Management
8.2.2 LSGDM Model Considering the Non-cooperative Behavior
8.3 Social Network LSGDM Considering Voluntary Trust Loss
8.3.1 Trust Cop-K-Means Clustering Analysis
8.3.2 Improved Minimum-Cost Consensus Model Considering Voluntary Trust Loss
8.3.3 Discussions and Comparative Analysis
References
9 New Perspectives on Consensus Reaching Process in Large-Scale Group Decision-Making
9.1 An LSGDM Method Balancing the Conflict Between CRP and DCAP
9.1.1 The Dynamic Clustering Analysis Process in LSGDM
9.1.2 The CRP in LSGDM Based on the DCAP
9.1.3 The Solution for LSGDM Balancing the DCAP and CRP
9.1.4 Case Study
9.2 A New Clustering Analysis Reducing the Complexity of CRP
9.2.1 Motivation on the Clustering Analysis Considering Adjustment Costs
9.2.2 A New Clustering Algorithm Considering Adjustment Costs
9.2.3 Main Procedures of the Clustering Analysis for LSGDM
9.2.4 A Case of Team Construction
References
10 Application of LSGDM Methods
10.1 LSGDM Model in Online-Customer Segmentation
10.1.1 Online Evaluation Process
10.1.2 Dynamic Online-Customer Segmentation with Semantic Evaluation Information
10.2 A Social Commerce Purchasing Model with Trust Network and Item Review Information
10.2.1 Motivation of Combining Trust Relationship and Item Review
10.2.2 S-Commerce Purchasing Decision Model Based on User Trust and Item Review
10.2.3 Inexperienced Purchasing Decision Based on the Composite Network Model
10.2.4 Application
10.3 LSGDM Model in Group Recommendation
10.3.1 Data Selection
10.3.2 Trust Network Construction with the “MovieLens-1 M” Dataset
10.3.3 The Application of the Proposed Solution
10.3.4 Comparison Analysis
10.4 Social Recommendation with LSGDM for Cyber-Enabled Online Service
10.4.1 Modeling of Scholarly Large-Scale Group Decision-Making Problem
10.4.2 A Two-Stage LSGDM Solution for Social Recommendation
10.4.3 Experiment and Analysis
References
Concluding Remarks
