Elsevier S & T

Pattern Recognition

Theodoridis, Sergios; Theodoridis, Sergios; Koutroumbas, Konstantinos; Koutroumbas, Konstantinos

In stock
No reviews
33.250 KD inc. VAT
License
Table of contents
  • Table of contentsv
  • PREFACExv
  • 1 INTRODUCTION1
  • 1.1 IS PATTERN RECOGNITION IMPORTANT?1
  • 1.2 FEATURES, FEATURE VECTORS, AND CLASSIFIERS3
  • 1.3 SUPERVISED VERSUS UNSUPERVISED PATTERN RECOGNITION6
  • 1.4 OUTLINE OF THE BOOK8
  • 2 CLASSIFIERS BASED ON BAYES DECISION THEORY13
  • 2.1 INTRODUCTION13
  • 2.2 BAYES DECISION THEORY13
  • 2.3 DISCRIMINANT FUNCTIONS AND DECISION SURFACES19
  • 2.4 BAYESIAN CLASSIFICATION FOR NORMAL DISTRIBUTIONS20
  • 2.5 ESTIMATION OF UNKNOWN PROBABILITY DENSITY FUNCTIONS28
  • 2.6 THE NEAREST NEIGHBOR RULE48
  • 2.7 BAYESIAN NETWORKS50
  • 3 LINEAR CLASSIFIERS69
  • 3.1 INTRODUCTION69
  • 3.2 LINEAR DISCRIMINANT FUNCTIONS AND DECISION HYPERPLANES69
  • 3.3 THE PERCEPTRON ALGORITHM71
  • 3.4 LEAST SQUARES METHODS79
  • 3.5 MEAN SQUARE ESTIMATION REVISITED86
  • 3.6 LOGISTIC DISCRIMINATION91
  • 3.7 SUPPORT VECTOR MACHINES93
  • 4 NONLINEAR CLASSIFIERS121
  • 4.1 INTRODUCTION121
  • 4.2 THE XOR PROBLEM121
  • 4.3 THE TWO-LAYER PERCEPTRON122
  • 4.4 THREE-LAYER PERCEPTRONS129
  • 4.5 ALGORITHMS BASED ON EXACT CLASSIFICATION OF THE TRAINING SET130
  • 4.6 THE BACKPROPAGATION ALGORITHM132
  • 4.7 VARIATIONS ON THE BACKPROPAGATION THEME140
  • 4.8 THE COST FUNCTION CHOICE143
  • 4.9 CHOICE OF THE NETWORK SIZE147
  • 4.10 A SIMULATION EXAMPLE153
  • 4.11 NETWORKS WITH WEIGHT SHARING155
  • 4.12 GENERALIZED LINEAR CLASSIFIERS156
  • 4.13 CAPACITY OF THE l-DIMENSIONAL SPACE IN LINEAR DICHOTOMIES158
  • 4.14 POLYNOMIAL CLASSIFIERS161
  • 4.15 RADIAL BASIS FUNCTION NETWORKS162
  • 4.16 UNIVERSAL APPROXIMATORS167
  • 4.17 SUPPORT VECTOR MACHINES: THE NONLINEAR CASE169
  • 4.18 DECISION TREES174
  • 4.19 COMBINING CLASSIFIERS181
  • 4.20 THE BOOSTING APPROACH TO COMBINE CLASSIFIERS188
  • 4.21 DISCUSSION196
  • 5 FEATURE SELECTION213
  • 5.1 INTRODUCTION213
  • 5.2 PREPROCESSING214
  • 5.3 FEATURE SELECTION BASED ON STATISTICAL HYPOTHESIS TESTING216
  • 5.4 THE RECEIVER OPERATING CHARACTERISTICS (ROC) CURVE223
  • 5.5 CLASS SEPARABILITY MEASURES224
  • 5.6 FEATURE SUBSET SELECTION231
  • 5.7 OPTIMAL FEATURE GENERATION237
  • 5.8 NEURAL NETWORKS AND FEATURE GENERATION/ SELECTION242
  • 5.9 A HINT ON GENERALIZATION THEORY243
  • 5.10 THE BAYESIAN INFORMATION CRITERION253
  • 6 FEATURE GENERATION I: LINEAR TRANSFORMS263
  • 6.1 INTRODUCTION263
  • 6.2 BASIS VECTORS AND IMAGES264
  • 6.3 THE KARHUNEN–LOÈVE TRANSFORM266
  • 6.4 THE SINGULAR VALUE DECOMPOSITION273
  • 6.5 INDEPENDENT COMPONENT ANALYSIS276
  • 6.6 THE DISCRETE FOURIER TRANSFORM (DFT)285
  • 6.7 THE DISCRETE COSINE AND SINE TRANSFORMS288
  • 6.8 THE HADAMARD TRANSFORM290
  • 6.9 THE HAAR TRANSFORM291
  • 6.10 THE HAAR EXPANSION REVISITED292
  • 6.11 DISCRETE TIMEWAVELET TRANSFORM (DTWT)297
  • 6.12 THE MULTIRESOLUTION INTERPRETATION307
  • 6.13 WAVELET PACKETS309
  • 6.14 A LOOK AT TWO-DIMENSIONAL GENERALIZATIONS311
  • 6.15 APPLICATIONS313
  • 7 FEATURE GENERATION II327
  • 7.1 INTRODUCTION327
  • 7.2 REGIONAL FEATURES328
  • 7.3 FEATURES FOR SHAPE AND SIZE CHARACTERIZATION353
  • 7.4 A GLIMPSE AT FRACTALS362
  • 7.5 TYPICAL FEATURES FOR SPEECH AND AUDIO CLASSIFICATION370
  • 8 TEMPLATE MATCHING397
  • 8.1 INTRODUCTION397
  • 8.2 MEASURES BASED ON OPTIMAL PATH SEARCHING TECHNIQUES398
  • 8.3 MEASURES BASED ON CORRELATIONS413
  • 8.4 DEFORMABLE TEMPLATE MODELS419
  • 9 CONTEXT-DEPENDENT CLASSIFICATION427
  • 9.1 INTRODUCTION427
  • 9.2 THE BAYES CLASSIFIER427
  • 9.3 MARKOV CHAIN MODELS428
  • 9.4 THE VITERBI ALGORITHM429
  • 9.5 CHANNEL EQUALIZATION432
  • 9.6 HIDDEN MARKOV MODELS437
  • 9.7 HMM WITH STATE DURATION MODELING452
  • 9.8 TRAINING MARKOV MODELS VIA NEURAL NETWORKS458
  • 9.9 A DISCUSSION OF MARKOV RANDOM FIELDS460
  • 10 SYSTEM EVALUATION471
  • 10.1 INTRODUCTION471
  • 10.2 ERROR COUNTING APPROACH471
  • 10.3 EXPLOITING THE FINITE SIZE OF THE DATA SET473
  • 10.4 A CASE STUDY FROM MEDICAL IMAGING476
  • 11 CLUSTERING: BASIC CONCEPTS483
  • 11.1 INTRODUCTION483
  • 11.2 PROXIMITY MEASURES490
  • 12 CLUSTERING ALGORITHMS I: SEQUENTIAL ALGORITHMS517
  • 12.1 INTRODUCTION517
  • 12.2 CATEGORIES OF CLUSTERING ALGORITHMS519
  • 12.3 SEQUENTIAL CLUSTERING ALGORITHMS523
  • 12.4 A MODIFICATION OF BSAS527
  • 12.5 A TWO-THRESHOLD SEQUENTIAL SCHEME529
  • 12.6 REFINEMENT STAGES531
  • 12.7 NEURAL NETWORK IMPLEMENTATION533
  • 13 CLUSTERING ALGORITHMS II: HIERARCHICAL ALGORITHMS541
  • 13.1 INTRODUCTION541
  • 13.2 AGGLOMERATIVE ALGORITHMS542
  • 13.3 THE COPHENETIC MATRIX568
  • 13.4 DIVISIVE ALGORITHMS570
  • 13.5 HIERARCHICAL ALGORITHMS FOR LARGE DATA SETS572
  • 13.6 CHOICE OF THE BEST NUMBER OF CLUSTERS580
  • 14 CLUSTERING ALGORITHMS III: SCHEMES BASED ON FUNCTION OPTIMIZATION589
  • 14.1 INTRODUCTION589
  • 14.2 MIXTURE DECOMPOSITION SCHEMES591
  • 14.3 FUZZY CLUSTERING ALGORITHMS600
  • 14.4 POSSIBILISTIC CLUSTERING622
  • 14.5 HARD CLUSTERING ALGORITHMS629
  • 14.6 VECTOR QUANTIZATION639
  • APPENDIX642
  • 15 CLUSTERING ALGORITHMS IV653
  • 15.1 INTRODUCTION653
  • 15.2 CLUSTERING ALGORITHMS BASED ON GRAPH THEORY653
  • 15.3 COMPETITIVE LEARNING ALGORITHMS660
  • 15.4 BINARY MORPHOLOGY CLUSTERING ALGORITHMS (BMCAs)669
  • 15.5 BOUNDARY DETECTION ALGORITHMS678
  • 15.6 VALLEY-SEEKING CLUSTERING ALGORITHMS681
  • 15.7 CLUSTERING VIA COST OPTIMIZATION (REVISITED)683
  • 15.8 KERNEL CLUSTERING METHODS692
  • 15.9 DENSITY-BASED ALGORITHMS FOR LARGE DATA SETS695
  • 15.10 CLUSTERING ALGORITHMS FOR HIGH-DIMENSIONAL DATA SETS702
  • 15.11 OTHER CLUSTERING ALGORITHMS718
  • 16 CLUSTER VALIDITY733
  • 16.1 INTRODUCTION733
  • 16.2 HYPOTHESIS TESTING REVISITED734
  • 16.3 HYPOTHESIS TESTING IN CLUSTER VALIDITY736
  • 16.4 RELATIVE CRITERIA747
  • 16.5 VALIDITY OF INDIVIDUAL CLUSTERS763
  • 16.6 CLUSTERING TENDENCY766
  • Appendix A HINTS FROM PROBABILITY AND STATISTICS785
  • A.1 TOTAL PROBABILITY AND THE BAYES RULE785
  • A.2 MEAN AND VARIANCE786
  • A.3 STATISTICAL INDEPENDENCE786
  • A.4 MARGINALIZATION786
  • A.5 CHARACTERISTIC FUNCTIONS787
  • A.6 MOMENTS AND CUMULANTS787
  • A.7 EDGEWORTH EXPANSION OF A PDF789
  • A.8 KULLBACK–LEIBLER DISTANCE789
  • A.9 MULTIVARIATE GAUSSIAN OR NORMAL PROBABILITY DENSITY FUNCTION790
  • A.10 THE CRAMER–RAO LOWER BOUND792
  • A.11 CENTRAL LIMIT THEOREM793
  • A.12 CHI-SQUARE DISTRIBUTION793
  • A.13 t-DISTRIBUTION795
  • A.14 BETA DISTRIBUTION795
  • A.15 POISSON DISTRIBUTION796
  • Appendix B LINEAR ALGEBRA BASICS797
  • B.1 POSITIVE DEFINITE AND SYMMETRIC MATRICES797
  • B.2 CORRELATION MATRIX DIAGONALIZATION798
  • Appendix C COST FUNCTION OPTIMIZATION801
  • C.1 GRADIENT DESCENT ALGORITHM801
  • C.2 NEWTON’S ALGORITHM805
  • C.3 CONJUGATE-GRADIENT METHOD806
  • C.4 OPTIMIZATION FOR CONSTRAINED PROBLEMS806
  • Appendix D BASIC DEFINITIONS FROM LINEAR SYSTEMS THEORY819
  • D.1 LINEAR TIME INVARIANT (LTI) SYSTEMS819
  • D.2 TRANSFER FUNCTION820
  • D.3 SERIAL AND PARALLEL CONNECTION821
  • D.4 TWO-DIMENSIONAL GENERALIZATIONS822
  • INDEX823
Book details
  • Vendor Elsevier S & T
  • SKU 9780123695314
  • ISBN-13 9780080513614
  • Author Theodoridis, Sergios; Theodoridis, Sergios; Koutroumbas, Konstantinos; Koutroumbas, Konstantinos
  • Edition 3rd
  • Category Computers
  • Subject Data Modeling & Design

Do you have questions about this book?

Ask an expert!

Description eBook delivery & access Reviews
Pattern recognition is a fast growing area with applications in a widely diverse number of fields such as communications engineering, bioinformatics, data mining, content-based database retrieval, to name but a few. This new edition addresses and keeps pace with the most recent advancements in these and related areas. This new edition: a) covers Data Mining, which was not treated in the previous edition, and is integrated with existing material in the book, b) includes new results on Learning Theory and Support Vector Machines, that are at the forefront of today's research, with a lot of interest both in academia and in applications-oriented communities, c) for the first time treats audio along with image applications since in today's world the most advanced applications are treated in a unified way and d) the subject of classifier combinations is treated, since this is a hot topic currently of interest in the pattern recognition community.

* The latest results on support vector machines including v-SVM's and their geometric interpretation
* Classifier combinations including the Boosting approach
* State-of-the-art material for clustering algorithms tailored for large data sets and/or high dimensional data, as required by applications such as web-mining and bioinformatics
* Coverage of diverse applications such as image analysis, optical character recognition, channel equalization, speech recognition and audio classification