Relational Data Clustering: Models, Algorithms, and Applications / Edition 1

Hardcover (Print)
Used and New from Other Sellers
Used and New from Other Sellers
from $80.01
Usually ships in 1-2 business days
(Save 12%)
Other sellers (Hardcover)
  • All (4) from $80.01   
  • New (2) from $80.01   
  • Used (2) from $117.87   

Overview

A culmination of the authors’ years of extensive research on this topic, Relational Data Clustering: Models, Algorithms, and Applications addresses the fundamentals and applications of relational data clustering. It describes theoretic models and algorithms and, through examples, shows how to apply these models and algorithms to solve real-world problems.

After defining the field, the book introduces different types of model formulations for relational data clustering, presents various algorithms for the corresponding models, and demonstrates applications of the models and algorithms through extensive experimental results. The authors cover six topics of relational data clustering:

  1. Clustering on bi-type heterogeneous relational data
  2. Multi-type heterogeneous relational data
  3. Homogeneous relational data clustering
  4. Clustering on the most general case of relational data
  5. Individual relational clustering framework
  6. Recent research on evolutionary clustering

This book focuses on both practical algorithm derivation and theoretical framework construction for relational data clustering. It provides a complete, self-contained introduction to advances in the field.

Read More Show Less

Product Details

Meet the Author

Bo Long is a scientist at Yahoo! Labs in Sunnyvale, California.

Zhongfei Zhang is an associate professor in the computer science department at the State University of New York in Binghamton.

Philip S. Yu is a professor in the computer science department and the Wexler Chair in Information Technology at the University of Illinois in Chicago.

Read More Show Less

Table of Contents

List of Tables

List of Figures

Preface

1 Introduction 1

1.1 Defining the Area 1

1.2 The Content and the Organization of This Book 4

1.3 The Audience of This Book 6

1.4 Further Readings 6

I Models 9

2 Co-Clustering 11

2.1 Introduction 11

2.2 Related Work 12

2.3 Model Formulation and Analysis 13

2.3.1 Block Value Decomposition 13

2.3.2 NBVD Method 17

3 Heterogeneous Relational Data Clustering 21

3.1 Introduction 21

3.2 Related Work 23

3.3 Relation Summary Network Model 24

4 Homogeneous Relational Data Clustering 29

4.1 Introduction 29

4.2 Related Work 32

4.3 Community Learning by Graph Approximation 33

5 General Relational Data Clustering 39

5.1 Introduction 39

5.2 Related Work 40

5.3 Mixed Membership Relational Clustering 42

5.4 Spectral Relational Clustering 45

6 Multiple-View Relational Data Clustering 47

6.1 Introduction 47

6.2 Related Work 49

6.3 Background and Model Formulation 50

6.3.1 A General Model for Multiple-View Unsupervised Learning 51

6.3.2 Two Specific Models: Multiple-View Clustering and Multiple-View Spectral Embedding 53

7 Evolutionary Data Clustering 57

7.1 Introduction 57

7.2 Related Work 59

7.3 Dirichlet Process Mixture Chain (DPChain) 60

7.3.1 DPChain Representation 61

7.4 HDP Evolutionary Clustering Model (HDP-EVO) 63

7.4.1 HDP-EVO Representation 63

7.4.2 Two-Level CRP for HDP-EVO 65

7.5 Infinite Hierarchical Hidden Markov State Model 66

7.5.1 iH2 MS Representation 67

7.5.2 Extention of iH2MS 68

7.5.3 Maximum Likelihood Estimation of HTM 69

7.6 HDP Incorporated with HTM (HDP-HTM) 70

7.6.1 Model Representation 70

II Algorithms 73

8 Co-Clustering 75

8.1 Nonnegative Block Value Decomposition Algorithm 75

8.2 Proof of the Correctness of the NBVD Algorithm 78

9 Heterogeneous Relational Data Clustering 83

9.1 Relation Summary Network Algorithm 83

9.2 A Unified View to Clustering Approaches 90

9.2.1 Bipartite Spectral Graph Partitioning 90

9.2.2 Binary Data Clustering with Feature Reduction 90

9.2.3 Information-Theoretic Co-Clustering 91

9.2.4 K-Means Clustering 92

10 Homogeneous Relational Data Clustering 95

10.1 Hard CLGA Algorithm 95

10.2 Soft CLGA Algorithm 97

10.3 Balanced CLGA Algorithm 101

11 General Relational Data Clustering 105

11.1 Mixed Membership Relational Clustering Algorithm 105

11.1.1 MMRC with Exponential Families 105

11.1.2 Monte Carlo E-Step 108

11.1.3 M-Step 109

11.1.4 Hard MMRC Algorithm 112

11.2 Spectral Relational Clustering Algorithm 114

11.3 A Unified View to Clustering 118

11.3.1 Semi-Supervised Clustering 118

11.3.2 Co-Clustering 119

11.3.3 Graph Clustering 120

12 Multiple-View Relational Data Clustering 123

12.1 Algorithm Derivation 123

12.1.1 Multiple-View Clustering Algorithm 124

12.1.2 Multiple-View Spectral Embedding Algorithm 127

12.2 Extensions and Discussions 129

12.2.1 Evolutionary Clustering 129

12.2.2 Unsupervised Learning with Side Information 130

13 Evolutionary Data Clustering 133

13.1 DPChain Inference 133

13.2 HDP-EVO Inference 134

13.3 HDP-HTM Inference 136

III Applications 139

14 Co-Clustering 141

14.1 Data Sets and Implementation Details 141

14.2 Evaluation Metricees 142

14.3 Results and Discussion 143

15 Heterogeneous Relational Data Clustering 147

15.1 Data Sets and Parameter Setting 147

15.2 Results and Discussion 150

16 Homogeneous Relational Data Clustering 153

16.1 Data Sets and Parameter Setting 153

16.2 Results and Discussion 155

17 General Relational Data Clustering 159

17.1 Graph Clustering 159

17.2 Bi-clustering and Tri-Clustering 161

17.3 A Case Study on Actor-Movie Data 163

17.4 Spectral Relational Clustering Applications 164

17.4.1 Clustering on Bi-Type Relational Data 164

17.4.2 Clustering on Tri-Type Relational Data 166

18 Multiple-View and Evolutionary Data Clustering 169

18.1 Multiple-View Clustering 169

18.1.1 Synthetic Data 169

18.1.2 Real Data 172

18.2 Multiple-View Spectral Embedding 173

18.3 Semi-Supervised Clustering 174

18.4 Evolutionary Clustering 175

IV Summary 179

References 185

Index 195

Read More Show Less

Customer Reviews

Be the first to write a review
( 0 )
Rating Distribution

5 Star

(0)

4 Star

(0)

3 Star

(0)

2 Star

(0)

1 Star

(0)

    If you find inappropriate content, please report it to Barnes & Noble
    Why is this product inappropriate?
    Comments (optional)