Support Vector Machines for Pattern ClassificationSpringer Science & Business Media, 2005 M12 28 - 344 pages I was shocked to see a student’s report on performance comparisons between support vector machines (SVMs) and fuzzy classi?ers that we had developed withourbestendeavors.Classi?cationperformanceofourfuzzyclassi?erswas comparable, but in most cases inferior, to that of support vector machines. This tendency was especially evident when the numbers of class data were small. I shifted my research e?orts from developing fuzzy classi?ers with high generalization ability to developing support vector machine–based classi?ers. This book focuses on the application of support vector machines to p- tern classi?cation. Speci?cally, we discuss the properties of support vector machines that are useful for pattern classi?cation applications, several m- ticlass models, and variants of support vector machines. To clarify their - plicability to real-world problems, we compare performance of most models discussed in the book using real-world benchmark data. Readers interested in the theoretical aspect of support vector machines should refer to books such as [109, 215, 256, 257]. |
Contents
| 2 | |
| 14 | |
6 | 150 |
5 | 166 |
Feature Selection and Extraction | 189 |
Clustering | 201 |
KernelBased Methods | 209 |
MaximumMargin Multilayer Neural Networks | 223 |
MaximumMargin Fuzzy Classifiers 237 | 236 |
Function Approximation | 265 |
A Conventional Classifiers | 297 |
Quadratic Programming | 309 |
References | 319 |
Index 339 | 338 |
Other editions - View all
Common terms and phrases
algorithm all-at-once Artificial Neural Networks belongs to class bias term blood cell data calculate class boundary cluster Conference on Neural convex hull cross-validation datum DDAG decision functions delete Di(x discuss dual problem ECOC exact KKT conditions feature space fuzzy classifier fuzzy support vector g(xi given Hessian matrix Hiragana-50 hyperplane hypersphere IEEE input space input variables KPCA L1 and L2 L2 support vector L2 SVM linear kernels m-dimensional Machine Learning Mahalanobis distance maximized membership functions multiclass Neural Networks IJCNN number of support number of training objective function obtain one-against-all support vector optimal hyperplane optimal solution orthogonal outliers output polynomial kernels positive definite quadratic programming RBF kernels recognition rates satisfied separating hyperplane set of support shown in Fig slack variables support vector machines support vector regressor Table test data Theorem thyroid data training data unbounded support vectors unclassifiable regions x1 Fig