| Description |
1 online resource |
| Series |
SpringerBriefs in complexity, 2191-5326 |
|
SpringerBriefs in complexity.
|
| Contents |
Preface; Acknowledgements; Contents; 1 Network Role Mining and Analysis: An Overview; 1.1 Introduction; 1.2 Defining Roles; 1.2.1 Networks; 1.2.2 Positions in Networks; 1.3 Mining Roles; 1.3.1 Relationship to Graph Partitioning and Community Detection; 1.4 Purpose and Outline of This Monograph; References; 2 Implied Role Mining; 2.1 Introduction; 2.2 The Implied Role Mining Process; 2.3 Illustrations with Usenet; 2.3.1 Golder et al.'s Taxonomy; 2.3.2 Nolker et al.'s Hierarchy; 2.4 Analysis of Implied Role Mining; 2.4.1 Qualitative Nature; 2.4.2 Compatibility |
|
2.4.3 Simplicity and Interpretability2.5 Conclusion; References; 3 Equivalence-Based Role Mining; 3.1 Introduction; 3.2 Structural Equivalence; 3.2.1 Finding Structural Equivalences; 3.3 Automorphic Equivalence; 3.3.1 Finding Automorphic Equivalences; 3.3.2 Quantifying Automorphic Similarity; 3.4 Regular Equivalence; 3.4.1 Finding Regular Equivalences; 3.4.2 Quantifying Regular Similarity; 3.5 Conclusion; References; 4 Deterministic Blockmodeling; 4.1 Introduction; 4.2 The Blockmodeling Framework; 4.2.1 Similarity Measures; 4.2.2 Blocktypes; 4.3 Goodness of Fit |
|
4.3.1 A Goodness-of-Fit Measure for Positional Analysis4.3.2 A Goodness-of-Fit Measure for Network Compression; 4.4 Conclusion; References; 5 Stochastic Blockmodeling; 5.1 Introduction; 5.2 SBM Specification; 5.3 The Infinite Relational Model; 5.3.1 Parameter Inference for the IRM; 5.3.2 Summary; 5.4 The Dynamic Stochastic Blockmodel; 5.4.1 DSBM Network Generation; 5.4.2 Parameter Inference for the DSBM; 5.5 Conclusion; References; 6 Advanced Computational Methods; 6.1 Factor Graphs: The Social Roles and Statuses Factor Graph Model; 6.1.1 Social Features; 6.1.2 A Factor Graph Model |
|
6.2 Multi-view Learning: Dual-View Uncertainty Regularization6.2.1 Graph Co-regularization; 6.2.2 Uncertainty Regularization and Objective Function; 6.3 Bayesian Modeling: Co-discovery of Roles in Communities; 6.4 Matrix Factorization: RolX; 6.5 Iterative Quadratic Programming: Synergistic Co-discovery #x83;; 6.5.1 Initializing Communities (InitCom); 6.5.2 Initializing Roles (InitRole); 6.5.3 Updating Communities; 6.5.4 Updating Roles; 6.6 Conclusion; References; 7 Concluding Remarks; 7.1 Emerging Trends in Role Mining; 7.2 Tension Between Rigor and Interpretability; References |
| Summary |
This brief presents readers with a summary of classic, modern, and state-of-the-art methods for discovering the roles of entities in networks (including social networks) that range from small to large-scale. It classifies methods by their mathematical underpinning, whether they are driven by implications about entity behaviors in system, or if they are purely data driven. The brief also discusses when and how each method should be applied, and discusses some outstanding challenges toward the development of future role mining methods of each type |
| Bibliography |
Includes bibliographical references |
| Notes |
Online resource; title from PDF title page (SpringerLink, viewed April 4, 2017) |
| Subject |
Data mining.
|
|
Data Mining
|
|
Mathematical physics.
|
|
Applied mathematics.
|
|
Business mathematics & systems.
|
|
The self, ego, identity, personality.
|
|
Data mining.
|
|
COMPUTERS -- General.
|
|
COMPUTERS -- Databases -- Data Mining.
|
|
Data mining
|
| Form |
Electronic book
|
| ISBN |
9783319538860 |
|
3319538861 |
|
