Graph-based social media analysis
| Titel: | Graph-based social media analysis / edited by Ioannis Pitas, Aristotle University of Thessaloniki Greece |
|---|---|
| Beteiligt: | |
| Veröffentlicht: | Boca Raton; London; New York : CRC Press, Taylor & Francis Group, [2016] |
| Umfang: | xvii, 424 Seiten : Illustrationen ; 26 cm |
| Format: | Buch |
| Sprache: | Englisch |
| Schriftenreihe/ mehrbändiges Werk: |
Chapman & Hall/CRC data mining and knowledge discovery series |
| ISBN: | 9781498719049 |
| Hinweise zum Inhalt: |
Inhaltsverzeichnis
Buch Bibliographie Register Cover Abstract |
X
- Preface
- p. xiii
- Contributors
- p. xv
- Editor Biography
- p. xvii
- 1
- Graphs in Social and Digital Media
- p. 1
- 1.1
- Introduction
- p. 1
- 1.2
- Dominant social networking/media platforms
- p. 3
- 1.3
- Collecting data from social media sites
- p. 5
- 1.4
- Social media graphs
- p. 8
- 1.4.1
- Graphs from Facebook data
- p. 8
- 1.4.2
- Graphs from Twitter data
- p. 10
- 1.4.3
- Graphs from bibliographic data
- p. 12
- 1.5
- Graph storage formats and visualization
- p. 14
- 1.6
- Big data issues in social and digital media
- p. 15
- 1.7
- Distributed computing platforms
- p. 15
- 1.8
- Conclusions
- p. 18
- Bibliography
- p. 18
- 2
- Mathematical Preliminaries: Graphs and Matrices
- p. 21
- 2.1
- Graph basics
- p. 21
- 2.2
- Linear algebra tools
- p. 24
- 2.3
- Matrix decompositions
- p. 28
- 2.4
- Vector and matrix derivatives
- p. 31
- Bibliography
- p. 31
- 3
- Algebraic Graph Analysis
- p. 35
- 3.1
- Introduction
- p. 35
- 3.2
- Spectral graph theory
- p. 36
- 3.2.1
- Adjacency and Laplacian matrix
- p. 36
- 3.2.2
- Similarity matrix and nearest neighbor graph
- p. 37
- 3.3
- Applications of graph analysis
- p. 38
- 3.4
- Random graph generation
- p. 40
- 3.4.1
- Desirable random graph properties
- p. 41
- 3.4.2
- Random graph generation models
- p. 41
- 3.4.3
- Spectral graph generation
- p. 43
- 3.5
- Graph clustering
- p. 45
- 3.5.1
- Global clustering algorithms
- p. 46
- 3.5.2
- Local clustering algorithms
- p. 48
- 3.5.3
- Spectral clustering algorithms
- p. 48
- 3.5.4
- Overlapping community detection
- p. 50
- 3.6
- Graph matching
- p. 51
- 3.6.1
- Spectral graph matching
- p. 53
- 3.6.2
- Frequent subgraph mining
- p. 54
- 3.7
- Random walks
- p. 54
- 3.8
- Graph anomaly detection
- p. 56
- 3.8.1
- Spectral anomaly detection
- p. 57
- 3.9
- Conclusions
- p. 58
- Bibliography
- p. 59
- 4
- Web Search Based on Ranking
- p. 67
- 4.1
- Introduction
- p. 67
- 4.2
- Information Retrieval Background
- p. 69
- 4.2.1
- Document representation
- p. 69
- 4.2.2
- Retrieval models
- p. 71
- 4.3
- Relevance Beyond the Web Page Text
- p. 72
- 4.3.1
- Anchor text
- p. 72
- 4.3.2
- Query expansion
- p. 73
- 4.4
- Centrality and Prestige
- p. 76
- 4.4.1
- Basic measures
- p. 77
- 4.4.2
- Eigenvector centrality and prestige
- p. 80
- 4.4.3
- PageRank
- p. 81
- 4.4.4
- Hubs and authorities
- p. 84
- 4.4.5
- SimRank
- p. 87
- 4.5
- Topic-Sensitive Ranking
- p. 88
- 4.5.1
- Content as topic
- p. 89
- 4.5.2
- Trust as topic
- p. 91
- 4.6
- Ranking in Heterogeneous Networks
- p. 92
- 4.6.1
- Ranking in heterogeneous information networks
- p. 93
- 4.6.2
- Ranking-Based clustering
- p. 95
- 4.7
- Organizing Search Results
- p. 97
- 4.8
- Conclusion
- p. 99
- Bibliography
- p. 100
- 5
- Label Propagation and Information Diffusion in Graphs
- p. 107
- 5.1
- Introduction
- p. 108
- 5.2
- Graph construction approaches
- p. 109
- 5.2.1
- Neighborhood approaches
- p. 110
- 5.2.2
- Local reconstruction approaches
- p. 111
- 5.2.3
- Metric learning approaches
- p. 113
- 5.2.4
- Scalable graph construction methods
- p. 118
- 5.3
- Label inference methods
- p. 120
- 5.3.1
- Iterative algorithms
- p. 120
- 5.3.2
- Random walks
- p. 122
- 5.3.3
- Graph regularization
- p. 123
- 5.3.4
- Graph kernel regularization
- p. 127
- 5.3.5
- Inductive label inference
- p. 128
- 5.3.6
- Label propagation on data with multiple representations
- p. 129
- 5.3.7
- Label propagation on hypergraphs
- p. 131
- 5.3.8
- Label propagation initialization
- p. 132
- 5.3.9
- Applications in digital media
- p. 133
- 5.4
- Diffusion processes
- p. 134
- 5.4.1
- Diffusion in physics
- p. 134
- 5.4.2
- Diffusion in sociology
- p. 135
- 5.4.3
- Diffusion in social media
- p. 135
- 5.5
- Social network diffusion models
- p. 136
- 5.5.1
- Game theoretical diffusion models
- p. 137
- 5.5.2
- Epidemic diffusion models
- p. 137
- 5.5.3
- Threshold diffusion models
- p. 138
- 5.5.4
- Cascade diffusion models
- p. 139
- 5.5.5
- Influence maximization
- p. 140
- 5.5.6
- Cross-Media information diffusion
- p. 142
- 5.5.7
- Other applications of information diffusion
- p. 143
- 5.6
- Conclusions
- p. 145
- Bibliography
- p. 146
- 6
- Graph-Based Pattern Classification and Dimensionality Reduction
- p. 163
- 6.1
- Introduction
- p. 163
- 6.2
- Notations
- p. 164
- 6.3
- Unsupervised Methods
- p. 166
- 6.3.1
- Locality Preserving Projections
- p. 166
- 6.3.2
- Locally Linear Embedding
- p. 167
- 6.3.3
- ISOMAP
- p. 168
- 6.3.4
- Laplacian Embedding
- p. 168
- 6.3.5
- Diffusion Maps
- p. 168
- 6.4
- Supervised Methods
- p. 169
- 6.4.1
- Linear Discriminant Analysis
- p. 169
- 6.4.2
- Marginal Fisher Analysis
- p. 171
- 6.4.3
- Local Fisher Discriminant Analysis
- p. 171
- 6.4.4
- Graph Embedding
- p. 172
- 6.4.5
- Minimum Class Variance Extreme Learning Machine
- p. 173
- 6.4.6
- Minimum Class Variance Support Vector Machine
- p. 174
- 6.4.7
- Graph Embedded Support Vector Machines
- p. 174
- 6.5
- Semi-Supervised Methods
- p. 175
- 6.5.1
- Semi-Supervised Discriminant Analysis
- p. 176
- 6.5.2
- Laplacian Support Vector Machine
- p. 176
- 6.5.3
- Semi-Supervised Extreme Learning Machine
- p. 177
- 6.6
- Applications
- p. 177
- 6.7
- Conclusions
- p. 178
- Bibliography
- p. 179
- 7
- Matrix and Tensor Factorization with Reconimender System Applications
- p. 187
- 7.1
- Introduction
- p. 187
- 7.2
- Singular Value Decomposition on Matrices for Recommender Systems
- p. 189
- 7.2.1
- Applying the SVD and Preserving the Largest Singular Values
- p. 190
- 7.2.2
- Generating the Neighborhood of Users/Items
- p. 191
- 7.2.3
- Generating the Recommendation List
- p. 191
- 7.2.4
- Inserting a Test User in the c-dimensional Space
- p. 192
- 7.2.5
- Other Factorization Methods
- p. 192
- 7.3
- Higher Order Singular Value Decomposition (HOSVD) on Tensors
- p. 193
- 7.3.1
- From SVD to HOSVD
- p. 193
- 7.3.2
- HOSVD for Recommendations in Social Tagging Systems
- p. 196
- 7.3.3
- Handling the Sparsity Problem
- p. 200
- 7.3.4
- Inserting New Users, Tags, or Items
- p. 201
- 7.3.5
- Other Scalable Factorization Models
- p. 204
- 7.4
- A Real Geo-Social System-Based on HOSVD
- p. 205
- 7.4.1
- GeoSocialRec Website
- p. 205
- 7.4.2
- GeoSocialRec Database and Recommendation Engine
- p. 207
- 7.4.3
- Experiments
- p. 208
- 7.5
- Conclusion
- p. 210
- Bibliography
- p. 210
- 8
- Multimedia Social Search Based on Hypergraph Learning
- p. 215
- 8.1
- Introduction
- p. 215
- 8.2
- Hypergraphs
- p. 218
- 8.2.1
- Uniform hypergraphs
- p. 220
- 8.3
- Game-Theoretic approaches to uniform hypergraph clustering
- p. 223
- 8.4
- Spectral clustering for arbitrary hypergraphs
- p. 229
- 8.5
- Ranking on hypergraphs
- p. 238
- 8.5.1
- Enforcing structural constraints
- p. 239
- 8.5.2
- Learning hypcredge weights
- p. 241
- 8.6
- Applications
- p. 243
- 8.6.1
- High-order web link analysis
- p. 243
- 8.6.2
- Hypergraph matching for object recognition
- p. 247
- 8.6.3
- Music recommendation and personalized music tagging
- p. 249
- 8.6.4
- Simultaneous image tagging and geo-location prediction
- p. 251
- 8.6.5
- Social image search exploiting joint visual-textual information
- p. 254
- 8.6.6
- Annotation, classification, and tourism recommendation driven by probabilistic latent semantic analysis
- p. 256
- 8.7
- Big data: Randomized methods for matrix/hypermatrix decompositions
- p. 261
- 8.8
- Conclusions
- p. 265
- 8.9
- Acknowledgments
- p. 267
- Bibliography
- p. 267
- 9
- Graph Signal Processing in Social Media
- p. 275
- 9.1
- Motivation
- p. 275
- 9.2
- Graph signal processing (GSP)
- p. 277
- 9.2.1
- Basics of graph signal processing
- p. 277
- 9.2.2
- Spectral representation of graph signals
- p. 279
- 9.2.3
- Downsampling in graphs
- p. 280
- 9.2.4
- Graph wavelets and filterbanks
- p. 282
- 9.3
- Applications
- p. 282
- 9.3.1
- Information diffusion pattern analysis
- p. 282
- 9.3.2
- Interpolation in graphs
- p. 284
- 9.3.2.1
- Movie recommendation system
- p. 286
- 9.4
- Conclusions
- p. 289
- Bibliography
- p. 289
- 10
- Big Data Analytics for Social Networks
- p. 293
- 10.1
- Introduction
- p. 294
- 10.1.1
- Signal processing for big data
- p. 294
- 10.1.2
- Social network analytics problems
- p. 295
- 10.2
- Visualizing and reducing dimension in social nets
- p. 296
- 10.2.1
- Kernel-based graph embedding
- p. 296
- 10.2.2
- Centrality-constraints
- p. 298
- 10.2.3
- Numerical tests
- p. 300
- 10.2.4
- Visualization of dynamic social networks
- p. 300
- 10.3
- Inference and imputation on social graphs
- p. 303
- 10.3.1
- Distributed anomaly detection for social graphs
- p. 303
- 10.3.1.1
- Anomaly detection via sparse plus low-rank decomposition
- p. 303
- 10.3.1.2
- In-network processing algorithm
- p. 305
- 10.3.1.3
- Numerical tests
- p. 306
- 10.3.2
- Prediction from partially-observed network processes
- p. 307
- 10.3.2.1
- Semi-supervised prediction of network processes
- p. 308
- 10.3.2.2
- Data-driven dictionary learning
- p. 309
- 10.3.2.3
- Numerical tests
- p. 310
- 10.4
- Unveiling communities in social networks
- p. 311
- 10.4.1
- Big data spectral clustering
- p. 312
- 10.4.1.1
- Numerical tests
- p. 315
- 10.4.2
- Robust kernel PCA
- p. 316
- 10.4.2.1
- Numerical tests
- p. 319
- 10.5
- Topology tracking from information cascades
- p. 319
- 10.5.1
- Dynamic SEMs for tracking cascades
- p. 321
- 10.5.1.1
- Model and problem statement
- p. 322
- 10.5.1.2
- Exponentially-weighted least-squares estimator
- p. 323
- 10.5.2
- Topology tracking algorithm
- p. 324
- 10.5.2.1
- Accelerated convergence
- p. 326
- 10.5.3
- Real-Time operation
- p. 327
- 10.5.3.1
- Premature termination
- p. 327
- 10.5.3.2
- Stochastic gradient descent iterations
- p. 327
- 10.5.4
- Experiments on real data
- p. 328
- 10.6
- Conclusion
- p. 330
- 10.7
- Acknowledgments
- p. 330
- Bibliography
- p. 330
- 11
- Semantic Model Adaptation for Evolving Big Social Data
- p. 341
- 11.1
- Introduction to Social Data Evolution
- p. 341
- 11.2
- Latent Model Adaptation
- p. 343
- 11.2.1
- Incremental Latent Semantic Analysis
- p. 343
- 11.2.2
- Incremental Probabilistic Latent Semantic Analysis
- p. 346
- 11.2.3
- Incremental Latent Dirichlet Allocation
- p. 355
- 11.3
- Incremental Spectral Clustering
- p. 359
- 11.4
- Tensor Model Adaptation
- p. 362
- 11.4.1
- Basic Tensor Concepts
- p. 362
- 11.4.2
- Incremental Tensor Analysis
- p. 363
- 11.5
- Parallel and Distributed Approaches for Big Data Analysis
- p. 368
- 11.5.1
- Parallel Probabilistic Latent Semantic Analysis
- p. 368
- 11.5.2
- Parallel Latent Dirichlet Allocation
- p. 369
- 11.5.3
- Parallel Spectral Clustering
- p. 371
- 11.5.4
- Distributed Tensor Decomposition
- p. 373
- 11.6
- Applications to Evolving Social Data Analysis
- p. 375
- 11.6.1
- Incremental Label Propagation
- p. 375
- 11.6.2
- Incremental Graph Clustering in Dynamic Social Networks
- p. 376
- 11.7
- Conclusions
- p. 379
- Bibliography
- p. 381
- 12
- Big Graph Storage, Processing and Visualization
- p. 391
- 12.1
- Introduction
- p. 391
- 12.2
- Basic Notions
- p. 393
- 12.3
- Big Graph Data Storage
- p. 395
- 12.3.1
- DBMS Architectures
- p. 395
- 12.3.2
- Graph DBMSs
- p. 397
- 12.3.3
- Storing and indexing graph structures
- p. 399
- 12.4
- Graph Data Processing
- p. 401
- 12.4.1
- Querying graphs in relational DBMS
- p. 402
- 12.4.2
- Graph querying in Datalog
- p. 403
- 12.4.3
- Query languages in graph DBMS
- p. 403
- 12.5
- Graph Data Visualization
- p. 405
- 12.5.1
- Static graph visualization
- p. 406
- 12.5.2
- Dynamic graph visualization
- p. 408
- 12.6
- Conclusions
- p. 409
- Bibliography
- p. 411
- Index
- p. 417
