Multimedia data mining : a systematic introduction to concepts and theory

cover image

Where to find it

Information & Library Science Library

Call Number
QA76.575 .Z53 2009
Status
Available

Summary

Collecting the latest developments in the field, Multimedia Data Mining: A Systematic Introduction to Concepts and Theory defines multimedia data mining, its theory, and its applications. Two of the most active researchers in multimedia data mining explore how this young area has rapidly developed in recent years.

The book first discusses the theoretical foundations of multimedia data mining, presenting commonly used feature representation, knowledge representation, statistical learning, and soft computing techniques. It then provides application examples that showcase the great potential of multimedia data mining technologies. In this part, the authors show how to develop a semantic repository training method and a concept discovery method in an imagery database. They demonstrate how knowledge discovery helps achieve the goal of imagery annotation. The authors also describe an effective solution to large-scale video search, along with an application of audio data classification and categorization.

This novel, self-contained book examines how the merging of multimedia and data mining research can promote the understanding and advance the development of knowledge discovery in multimedia data.

Contents

  • I Introduction p. 1
  • 1 Introduction p. 3
  • 1.1 Defining the Area p. 3
  • 1.2 A Typical Architecture of a Multimedia Data Mining System p. 7
  • 1.3 The Content and the Organization of This Book p. 8
  • 1.4 The Audience of This Book p. 10
  • 1.5 Further Readings p. 11
  • II Theory and Techniques p. 13
  • 2 Feature and Knowledge Representation for Multimedia Data p. 15
  • 2.1 Introduction p. 15
  • 2.2 Basic Concepts p. 16
  • 2.2.1 Digital Sampling p. 17
  • 2.2.2 Media Types p. 18
  • 2.3 Feature Representation p. 22
  • 2.3.1 Statistical Features p. 23
  • 2.3.2 Geometric Features p. 29
  • 2.3.3 Meta Features p. 32
  • 2.4 Knowledge Representation p. 32
  • 2.4.1 Logic Representation p. 33
  • 2.4.2 Semantic Networks p. 34
  • 2.4.3 Frames p. 36
  • 2.4.4 Constraints p. 38
  • 2.4.5 Uncertainty Representation p. 41
  • 2.5 Summary p. 44
  • 3 Statistical Mining Theory and Techniques p. 45
  • 3.1 Introduction p. 45
  • 3.2 Bayesian Learning p. 47
  • 3.2.1 Bayes Theorem p. 47
  • 3.2.2 Bayes Optimal Classifier p. 49
  • 3.2.3 Gibbs Algorithm p. 50
  • 3.2.4 Naive Bayes Classifier p. 50
  • 3.2.5 Bayesian Belief Networks p. 52
  • 3.3 Probabilistic Latent Semantic Analysis p. 56
  • 3.3.1 Latent Semantic Analysis p. 57
  • 3.3.2 Probabilistic Extension to Latent Semantic Analysis p. 58
  • 3.3.3 Model Fitting with the EM Algorithm p. 60
  • 3.3.4 Latent Probability Space and Probabilistic Latent Semantic Analysis p. 61
  • 3.3.5 Model Overfitting and Tempered EM p. 62
  • 3.4 Latent Dirichlet Allocation for Discrete Data Analysis p. 63
  • 3.4.1 Latent Dirichlet Allocation p. 64
  • 3.4.2 Relationship to Other Latent Variable Models p. 66
  • 3.4.3 Inference in LDA p. 69
  • 3.4.4 Parameter Estimation in LDA p. 70
  • 3.5 Hierarchical Dirichlet Process p. 72
  • 3.6 Applications in Multimedia Data Mining p. 73
  • 3.7 Support Vector Machines p. 74
  • 3.8 Maximum Margin Learning for Structured Output Space p. 81
  • 3.9 Boosting p. 88
  • 3.10 Multiple Instance Learning p. 91
  • 3.10.1 Establish the Mapping between the Word Space and the Image-VRep Space p. 93
  • 3.10.2 Word-to-Image Querying p. 95
  • 3.10.3 Image-to-Image Querying p. 95
  • 3.10.4 Image-to-Word Querying p. 96
  • 3.10.5 Multimodal Querying p. 96
  • 3.10.6 Scalability Analysis p. 97
  • 3.10.7 Adaptability Analysis p. 97
  • 3.11 Semi-Supervised Learning p. 101
  • 3.11.1 Supervised Learning p. 104
  • 3.11.2 Semi-Supervised Learning p. 106
  • 3.11.3 Semiparametric Regularized Least Squares p. 109
  • 3.11.4 Semiparametric Regularized Support Vector Machines p. 111
  • 3.11.5 Semiparametric Regularization Algorithm p. 113
  • 3.11.6 Transductive Learning and Semi-Supervised Learning p. 113
  • 3.11.7 Comparisons with Other Methods p. 114
  • 3.12 Summary p. 115
  • 4 Soft Computing Based Theory and Techniques p. 117
  • 4.1 Introduction p. 117
  • 4.2 Characteristics of the Paradigms of Soft Computing p. 118
  • 4.3 Fuzzy Set Theory p. 119
  • 4.3.1 Basic Concepts and Properties of Fuzzy Sets p. 119
  • 4.3.2 Fuzzy Logic and Fuzzy Inference Rules p. 123
  • 4.3.3 Fuzzy Set Application in Multimedia Data Mining p. 124
  • 4.4 Artificial Neural Networks p. 125
  • 4.4.1 Basic Architectures of Neural Networks p. 125
  • 4.4.2 Supervised Learning in Neural Networks p. 131
  • 4.4.3 Reinforcement Learning in Neural Networks p. 136
  • 4.5 Genetic Algorithms p. 140
  • 4.5.1 Genetic Algorithms in a Nutshell p. 140
  • 4.5.2 Comparison of Conventional and Genetic Algorithms for an Extremum Search p. 145
  • 4.6 Summary p. 150
  • III Multimedia Data Mining Application Examples p. 153
  • 5 Image Database Modeling - Semantic Repository Training p. 155
  • 5.1 Introduction p. 155
  • 5.2 Background p. 156
  • 5.3 Related Work p. 157
  • 5.4 Image Features and Visual Dictionaries p. 159
  • 5.4.1 Image Features p. 159
  • 5.4.2 Visual Dictionary p. 160
  • 5.5 [alpha]-Semantics Graph and Fuzzy Model for Repositories p. 163
  • 5.5.1 [alpha]-Semantics Graph p. 163
  • 5.5.2 Fuzzy Model for Repositories p. 166
  • 5.6 Classification Based Retrieval Algorithm p. 168
  • 5.7 Experiment Results p. 170
  • 5.7.1 Classification Performance on a Controlled Database p. 170
  • 5.7.2 Classification Based Retrieval Results p. 172
  • 5.8 Summary p. 180
  • 6 Image Database Modeling - Latent Semantic Concept Discovery p. 181
  • 6.1 Introduction p. 181
  • 6.2 Background and Related Work p. 182
  • 6.3 Region Based Image Representation p. 185
  • 6.3.1 Image Segmentation p. 185
  • 6.3.2 Visual Token Catalog p. 188
  • 6.4 Probabilistic Hidden Semantic Model p. 191
  • 6.4.1 Probabilistic Database Model p. 191
  • 6.4.2 Model Fitting with EM p. 192
  • 6.4.3 Estimating the Number of Concepts p. 194
  • 6.5 Posterior Probability Based Image Mining and Retrieval p. 194
  • 6.6 Approach Analysis p. 196
  • 6.7 Experimental Results p. 199
  • 6.8 Summary p. 205
  • 7 A Multimodal Approach to Image Data Mining and Concept Discovery p. 209
  • 7.1 Introduction p. 209
  • 7.2 Background p. 210
  • 7.3 Related Work p. 211
  • 7.4 Probabilistic Semantic Model p. 213
  • 7.4.1 Probabilistically Annotated Image Model p. 213
  • 7.4.2 EM Based Procedure for Model Fitting p. 215
  • 7.4.3 Estimating the Number of Concepts p. 216
  • 7.5 Model Based Image Annotation and Multimodal Image Mining and Retrieval p. 217
  • 7.5.1 Image Annotation and Image-to-Text Querying p. 217
  • 7.5.2 Text-to-Image Querying p. 218
  • 7.6 Experiments p. 219
  • 7.6.1 Dataset and Feature Sets p. 220
  • 7.6.2 Evaluation Metrics p. 221
  • 7.6.3 Results of Automatic Image Annotation p. 221
  • 7.6.4 Results of Single Word Text-to-Image Querying p. 224
  • 7.6.5 Results of Image-to-Image Querying p. 224
  • 7.6.6 Results of Performance Comparisons with Pure Text Indexing Methods p. 226
  • 7.7 Summary p. 228
  • 8 Concept Discovery and Mining in a Video Database p. 231
  • 8.1 Introduction p. 231
  • 8.2 Background p. 232
  • 8.3 Related Work p. 233
  • 8.4 Video Categorization p. 235
  • 8.4.1 Naive Bayes Classifier p. 237
  • 8.4.2 Maximum Entropy Classifier p. 238
  • 8.4.3 Support Vector Machine Classifier p. 240
  • 8.4.4 Combination of Meta Data and Content Based Classifiers p. 241
  • 8.5 Query Categorization p. 242
  • 8.6 Experiments p. 244
  • 8.6.1 Data Sets p. 244
  • 8.6.2 Video Categorization Results p. 246
  • 8.6.3 Query Categorization Results p. 251
  • 8.6.4 Search Relevance Results p. 253
  • 8.7 Summary p. 255
  • 9 Concept Discovery and Mining in an Audio Database p. 257
  • 9.1 Introduction p. 257
  • 9.2 Background and Related Work p. 258
  • 9.3 Feature Extraction p. 260
  • 9.4 Classification Method p. 263
  • 9.5 Experimental Results p. 263
  • 9.6 Summary p. 269
  • References p. 271
  • Index p. 291

Subjects

Subject Headings A:

Other details