Rough-Fuzzy Pattern Recognition Applications in Bioinformatics and Medical Imaging

by Maji, Pradipta; Pal, Sankar K.

• ISBN13: 9781118004401
• ISBN10: 111800440X
• Edition: 1st
• Format: Hardcover
• Copyright: 2012-02-14
• Publisher: Wiley-IEEE Computer Society Pr

Summary

This book provides a unified framework describing how rough-fuzzy computing techniques can be formulated and used in building efficient pattern recognition models. Based on the existing as well as new results, the book is structured according to the major phases of a pattern recognition system (e.g., classification, clustering, and feature selection) with a balanced mixture of theory, algorithm and applications. Special emphasis has been given to applications in bioinformatics and medical image processing. The book is useful for graduate students and researchers in computer science, electrical engineering, system science, medical science, and information technology. Researchers and practitioners in industry and R&D laboratories will also benefit.

Author Biography

Pradipta Maji, PhD, is Assistant Professor in the Machine Intelligence Unit of the Indian Statistical Institute. His research explores pattern recognition, bioinformatics, medical image processing, cellular automata, and soft computing. Sankar K. Pal, PhD, is Director and Distinguished Scientist of the Indian Statistical Institute. He is also a J. C. Bose Fellow of the Government of India. Dr. Pal founded both the Machine Intelligence Unit and the Center for Soft Computing Research at the Indian Statistical Institute. He is a Fellow of the IEEE, IAPR, IFSA, TWAS, and Indian National Science Academy.

Table of Contents

Foreword	p. xiii
Preface	p. xv
About the Authors	p. xix
Introduction to Pattern Recognition and Data Mining	p. 1
Introduction	p. 1
Pattern Recognition	p. 3
Data Acquisition	p. 4
Feature Selection	p. 4
Classification and Clustering	p. 5
Data Mining	p. 6
Tasks, Tools, and Applications	p. 7
Pattern Recognition Perspective	p. 8
Relevance of Soft Computing	p. 9
Scope and Organization of the Book	p. 10
References	p. 14
Rough-Fuzzy Hybridization and Granular Computing	p. 21
Introduction	p. 21
Fuzzy Sets	p. 22
Rough Sets	p. 23
Emergence of Rough-Fuzzy Computing	p. 26
Granular Computing	p. 26
Computational Theory of Perception and f-Granulation	p. 26
Rough-Fuzzy Computing	p. 28
Generalized Rough Sets	p. 29
Entropy Measures	p. 30
Conclusion and Discussion	p. 36
References	p. 37
Rough-Fuzzy Clustering: Generalized c-Means Algorithm	p. 47
Introduction	p. 47
Existing c-Means Algorithms	p. 49
Hard c-Means	p. 49
Fuzzy c-Means	p. 50
Possibilistic c-Means	p. 51
Rough c-Means	p. 52
Rough-Fuzzy-Possibilistic c-Means	p. 53
Objective Function	p. 54
Cluster Prototypes	p. 55
Fundamental Properties	p. 56
Convergence Condition	p. 57
Details of the Algorithm	p. 59
Selection of Parameters	p. 60
Generalization of Existing c-Means Algorithms	p. 61
RFCM: Rough-Fuzzy c-Means	p. 61
RPCM: Rough-Possibilistic c-Means	p. 62
RCM: Rough c-Means	p. 63
FPCM: Fuzzy-Possibilistic c-Means	p. 64
FCM: Fuzzy c-Means	p. 64
PCM: Possibilistic c-Means	p. 64
HCM: Hard c-Means	p. 65
Quantitative Indices for Rough-Fuzzy Clustering	p. 65
Average Accuracy, ¿ Index	p. 65
Average Roughness, ¿ Index	p. 67
Accuracy of Approximation, ¿* Index	p. 67
Quality of Approximation, ¿ Index	p. 68
Performance Analysis	p. 68
Quantitative Indices	p. 68
Synthetic Data Set: X32	p. 69
Benchmark Data Sets	p. 70
Conclusion and Discussion	p. 80
References	p. 81
Rough-Fuzzy Granulation and Pattern Classification	p. 85
Introduction	p. 85
Pattern Classification Model	p. 87
Class-Dependent Fuzzy Granulation	p. 88
Rough-Set-Based Feature Selection	p. 90
Quantitative Measures	p. 95
Dispersion Measure	p. 95
Classification Accuracy, Precision, and Recall	p. 96
¿ Coefficient	p. 96
ß Index	p. 97
Description of Data Sets	p. 97
Completely Labeled Data Sets	p. 98
Partially Labeled Data Sets	p. 99
Experimental Results	p. 100
Statistical Significance Test	p. 102
Class Prediction Methods	p. 103
Performance on Completely Labeled Data	p. 103
Performance on Partially Labeled Data	p. 110
Conclusion and Discussion	p. 112
References	p. 114
Fuzzy-Rough Feature Selection using f-Information Measures	p. 117
Introduction	p. 117
Fuzzy-Rough Sets	p. 120
Information Measure on Fuzzy Approximation Spaces	p. 121
Fuzzy Equivalence Partition Matrix and Entropy	p. 121
Mutual Information	p. 123
f-Information and Fuzzy Approximation Spaces	p. 125
V-Information	p. 125
I¿-Information	p. 126
M¿-Information	p. 127
¿¿-Information	p. 127
Hellinger Integral	p. 128
Renyi Distance	p. 128
f-Information for Feature Selection	p. 129
Feature Selection Using f-Information	p. 129
Computational Complexity	p. 130
Fuzzy Equivalence Classes	p. 131
Quantitative Measures	p. 133
Fuzzy-Rough-Set-Based Quantitative Indices	p. 133
Existing Feature Evaluation Indices	p. 133
Experimental Results	p. 135
Description of Data Sets	p. 136
Illustrative Example	p. 137
Effectiveness of the FEPM-Based Method	p. 138
Optimum Value of Weight Parameter ß	p. 141
Optimum Value of Multiplicative Parameter ¿	p. 141
Performance of Different f-Information Measures	p. 145
Comparative Performance of Different Algorithms	p. 152
Conclusion and Discussion	p. 156
References	p. 156
Rough Fuzzy c-Medoids and Amino Acid Sequence Analysis	p. 161
Introduction	p. 161
Bio-Basis Function and String Selection Methods	p. 164
Bio-Basis Function	p. 164
Selection of Bio-Basis Strings Using Mutual Information	p. 166
Selection of Bio-Basis Strings Using Fisher Ratio	p. 167
Fuzzy-Possibilistic c-Medoids Algorithm	p. 168
Hard c-Medoids	p. 168
Fuzzy c-Medoids	p. 169
Possibilistic c-Medoids	p. 170
Fuzzy-Possibilistic c-Medoids	p. 171
Rough-Fuzzy c-Medoids Algorithm	p. 172
Rough c-Medoids	p. 172
Rough-Fuzzy c-Medoids	p. 174
Relational Clustering for Bio-Basis String Selection	p. 176
Quantitative Measures	p. 178
Using Homology Alignment Score	p. 178
Using Mutual Information	p. 179
Experimental Results	p. 181
Description of Data Sets	p. 181
Illustrative Example	p. 183
Performance Analysis	p. 184
Conclusion and Discussion	p. 196
References	p. 196
Clustering Functionally Similar Genes from Microarray Data	p. 201
Introduction	p. 201
Clustering Gene Expression Data	p. 203
it-Means Algorithm	p. 203
Self-Organizing Map	p. 203
Hierarchical Clustering	p. 204
Graph-Theoretical Approach	p. 204
Model-Based Clustering	p. 205
Density-Based Hierarchical Approach	p. 206
Fuzzy Clustering	p. 206
Rough-Fuzzy Clustering	p. 206
Quantitative and Qualitative Analysis	p. 207
Silhouette Index	p. 207
Eisen and Cluster Profile Plots	p. 207
Z Score	p. 208
Gene-Ontology-Based Analysis	p. 208
Description of Data Sets	p. 209
Fifteen Yeast Data	p. 209
Yeast Sporulation	p. 211
Auble Data	p. 211
Cho et al. Data	p. 211
Reduced Cell Cycle Data	p. 211
Experimental Results	p. 212
Performance Analysis of Rough-Fuzzy c-Means	p. 212
Comparative Analysis of Different c-Means	p. 212
Biological Significance Analysis	p. 215
Comparative Analysis of Different Algorithms	p. 215
Performance Analysis of Rough-Fuzzy-Possibilistic c-Means	p. 217
Conclusion and Discussion	p. 217
References	p. 220
Selection of Discriminative Genes from Microarray Data	p. 225
Introduction	p. 225
Evaluation Criteria for Gene Selection	p. 227
Statistical Tests	p. 228
Euclidean Distance	p. 228
Pearson's Correlation	p. 229
Mutual Information	p. 229
f-Information Measures	p. 230
Approximation of Density Function	p. 230
Discretization	p. 231
Parzen Window Density Estimator	p. 231
Fuzzy Equivalence Partition Matrix	p. 233
Gene Selection using Information Measures	p. 234
Experimental Results	p. 235
Support Vector Machine	p. 235
Gene Expression Data Sets	p. 236
Performance Analysis of the FEPM	p. 236
Comparative Performance Analysis	p. 250
Conclusion and Discussion	p. 250
References	p. 252
Segmentation of Brain Magnetic Resonance Images	p. 257
Introduction	p. 257
Pixel Classification of Brain MR Images	p. 259
Performance on Real Brain MR Images	p. 260
Performance on Simulated Brain MR Images	p. 263
Segmentation of Brain MR Images	p. 264
Feature Extraction	p. 265
Selection of Initial Prototypes	p. 274
Experimental Results	p. 277
Illustrative Example	p. 277
Importance of Homogeneity and Edge Value	p. 278
Importance of Discriminant Analysis-Based Initialization	p. 279
Comparative Performance Analysis	p. 280
Conclusion and Discussion	p. 283
References	p. 283
Index	p. 287
Table of Contents provided by Ingram. All Rights Reserved.