Support Vector Machines (SVM technology) is one of the most user-friendly learning technologies available. Knowledge Discovery with Support Vector Machines provides an accessible introduction to model building and knowledge discovery with one of the preeminent algorithms. By introducing the reader to the subject material in a well-motivated setting where the technical prerequisites are kept to a minimum, the book presents a gentle tutorial on support vector machines for software engineers, software architects, data miners, bioinformatics specialists, analysts, statisticians, and other practitioners.

Lutz Hamel, PhD, teaches at the University of Rhode Island, where he founded the machine learning and data mining group. His major research interests are computational logic, machine learning, evolutionary computation, data mining, bioinformatics, and computational structures in art?and literature.

Preface	p. xiii
	p. 1
What Is Knowledge Discovery?	p. 3
Machine Learning	p. 4
Structure of the Universe X	p. 6
Inductive Learning	p. 8
Model Representations	p. 9
Exercises	p. 11
Bibliographic Notes	p. 11
Knowledge Discovery Environments	p. 13
Computational Aspects of Knowledge Discovery	p. 13
Data Access	p. 14
Visualization	p. 17
Data Manipulation	p. 20
Model Building and Evaluation	p. 23
Model Deployment	p. 26
Other Tool Sets	p. 27
Exercises	p. 27
Bibliographic Notes	p. 28
Describing Data Mathematically	p. 31
From Data Sets to Vector Spaces	p. 31
Vectors	p. 35
Vector Spaces	p. 40
The Dot Product as a Similarity Score	p. 41
Lines, Planes, and Hyperplanes	p. 44
Exercises	p. 47
Bibliographic Notes	p. 48
Linear Decision Surfaces and Functions	p. 49
From Data Sets to Decision Functions	p. 49
Linear Decision Surfaces Through the Origin	p. 50
Decision Surfaces with an Offset Term	p. 51
Simple Learning Algorithm	p. 54
Discussion	p. 57
Exercises	p. 58
Bibliographic Notes	p. 59
Perceptron Learning	p. 61
Perceptron Architecture and Training	p. 62
Duality	p. 67
Discussion	p. 70
Exercises	p. 71
Bibliographic Notes	p. 72
Maximum-Margin Classifiers	p. 73
Optimization Problems	p. 74
Maximum Margins	p. 75
Optimizing the Margin	p. 77
Quadratic Programming	p. 82
Discussion	p. 86
Exercises	p. 87
Bibliographic Notes	p. 88
	p. 89
Support Vector Machines	p. 91
The Lagrangian Dual	p. 92
Dual Maximum-Margin Optimization	p. 97
The Dual Decision Function	p. 101
Linear Support Vector Machines	p. 102
Nonlinear Support Vector Machines	p. 103
The Kernel Trick	p. 106
Feature Search	p. 109
A Closer Look at Kernels	p. 109
Soft-Margin Classifiers	p. 114
The Dual Setting for Soft-Margin Classifiers	p. 118
Tool Support	p. 122
WEKA	p. 123
R	p. 126
Discussion	p. 128
Exercises	p. 130
Bibliographic Notes	p. 131
Implementation	p. 133
Gradient Ascent	p. 134
The Kernel-Adatron Algorithm	p. 136
Quadratic Programming	p. 138
Chunking	p. 139
Sequential Minimal Optimization	p. 142
Discussion	p. 144
Exercises	p. 144
Bibliographic Notes	p. 145
Evaluating What Has Been Learned	p. 147
Performance Metrics	p. 148
The Confusion Matrix	p. 149
Model Evaluation	p. 152
The Hold-Out Method	p. 155
The Leave-One-Out Method	p. 157
N-Fold Cross-Validation	p. 158
Error Confidence Intervals	p. 160
Comparison of Models	p. 162
Model Evaluation in Practice	p. 163
WEKA	p. 163
R	p. 167
Exercises	p. 169
Bibliographic Notes	p. 170
Elements Of Statistical Learning Theory	p. 171
The VC-Dimension and Model Complexity	p. 172
A Theoretical Setting for Machine Learning	p. 175
Empirical Risk Minimization	p. 176
VC-Confidence	p. 177
Structural Risk Minimization	p. 179
Discussion	p. 180
Exercises	p. 180
Bibliographic Notes	p. 181
	p. 183
Multiclass Classification	p. 185
One-Versus-the-Rest Classification	p. 185
Pairwise Classification	p. 189
Discussion	p. 192
Exercises	p. 192
Bibliographic Notes	p. 192
Regression With Support Vector Machines	p. 193
Regression as Machine Learning	p. 193
Simple and Multiple Linear Regression	p. 194
Regression with Maximum-Margin Machines	p. 197
Regression with Support Vector Machines	p. 200
Model Evaluation	p. 202
Toot Support	p. 203
WEKA	p. 204
R	p. 205
Exercises	p. 207
Bibliographic Notes	p. 208
Novelty Detection	p. 209
Maximum-Margin Machines	p. 210
The Dual Setting	p. 212
Novelty Detection in R	p. 214
Exercises	p. 217
Bibliographic Notes	p. 217
Notation	p. 219
Tutorial Introduction To R	p. 221
Programming Constructs	p. 222
Data Constructs	p. 224
Basic Data Analysis	p. 227
Bibliographic Notes	p. 230
References	p. 231
Index	p. 237
Table of Contents provided by Ingram. All Rights Reserved.

What is included with this book?

The New copy of this book will include any supplemental materials advertised. Please check the title of the book to determine if it should include any access cards, study guides, lab manuals, CDs, etc.

The Used, Rental and eBook copies of this book are not guaranteed to include any supplemental materials. Typically, only the book itself is included. This is true even if the title states it includes any access cards, study guides, lab manuals, CDs, etc.