Which Is the Best Multiclass SVM Method? An Empirical Study
نویسندگان
چکیده
Multiclass SVMs are usually implemented by combining several two-class SVMs. The one-versus-all method using winner-takes-all strategy and the one-versus-one method implemented by max-wins voting are popularly used for this purpose. In this paper we give empirical evidence to show that these methods are inferior to another one-versusone method: one that uses Platt’s posterior probabilities together with the pairwise coupling idea of Hastie and Tibshirani. The evidence is particularly strong when the training dataset is sparse.
منابع مشابه
A Comparison of SVM-based Evolutionary Methods for Multicategory Cancer Diagnosis using Microarray Gene Expression Data
Selection of relevant genes that will give higher accuracy for sample classification (for example, to distinguish cancerous from normal tissues) is a common task in most microarray data studies. An evolutionary method based on generalization error bound theory of support vector machine (SVM) can select a subset of potentially informative genes for SVM classifier very efficiently. The bound theo...
متن کاملIndividual feature selection in each One-versus-One classifier improves multi-class SVM performance
Multiclass One-versus-One (OvO) SVM, which is constructed by assembling a group of binary classifiers, is usually treated as a black-box. The usual Multiclass Feature Selection (MFS) algorithm chooses an identical subset of features for every OvO SVM. We question whether the standard process of applying feature selection and then constructing the multiclass classifier is best. We propose that I...
متن کاملMulticategory Classification by Support Vector Machines
We examine the problem of how to discriminate between objects of three or more classes. Specifically, we investigate how two-class discrimination methods can be extended to the multiclass case. We show how the linear programming (LP) approaches based on the work of Mangasarian and quadratic programming (QP) approaches based on Vapnik’s Support Vector Machines (SVM) can be combined to yield two ...
متن کاملUniversum Learning for Multiclass SVM
We introduce Universum learning [1], [2] for multiclass problems and propose a novel formulation for multiclass universum SVM (MU-SVM). We also propose a span bound for MU-SVM that can be used for model selection thereby avoiding resampling. Empirical results demonstrate the effectiveness of MU-SVM and the proposed bound.
متن کاملAn efficient SMO-like algorithm for multiclass SVM
Starting from a reformulation of Cramer & Singer Multiclass Kernel Machine, we propose a Sequential Minimal Optimization (SMO) like algorithm for incremental and fast optimization of the lagrangian. The proposed formulation allowed us to de ne very e ective new pattern selection strategies which lead to better empirical results.
متن کامل