Multiclass Classification with Pairwise Coupled Neural Networks or Support Vector Machines

Authors:
Eddy Mayoraz
Affiliations:
-
Venue:
ICANN '01 Proceedings of the International Conference on Artificial Neural Networks
Year:
2001

Citing 10
Cited 0

Links between Markov models and multilayer perceptrons

Advances in neural information processing systems 1
Pairwise classification and support vector machines

Advances in kernel methods
Approximate statistical tests for comparing supervised classification learning algorithms

Neural Computation
On the Decomposition of Polychotomies into Dichotomies

ICML '97 Proceedings of the Fourteenth International Conference on Machine Learning
Phone Classification with Segmental Features and a Binary-Pair Partitioned Neural Network Classifier

ICASSP '97 Proceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP '97)-Volume 2 - Volume 2
Telephone Based Speaker Recognition Using Multiple Binary Classifier and Gaussian Mixture Models

ICASSP '97 Proceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP '97)-Volume 2 - Volume 2
SVMTorch: support vector machines for large-scale regression problems

The Journal of Machine Learning Research
Combined 5 × 2 cv F Test for Comparing Supervised Classification Learning Algorithms

Neural Computation
Speaker identification via support vector classifiers

ICASSP '96 Proceedings of the Acoustics, Speech, and Signal Processing, 1996. on Conference Proceedings., 1996 IEEE International Conference - Volume 01
Review:

Neural Computation

Quantified Score

Hi-index	0.02

Visualization

Abstract

Support Vector Machines (SVMs) are traditionally used for multi-class classification by introducing for each class one SVM trained to distinguish the associated class from all the others. In a recent experiment, we attempted to solve a K-class problem using a similar decomposition with K feedforward binary neural networks. The disappointing results were explained by the fact that neural networks suffer from datasets with a strongly unbalanced class distribution. By opposition to one-per-class, pairwise coupling introduces one binary classifier for each pair of classes and does not degrade the original class distribution. A few papers report evidences that pairwise coupling gives better results for SVMs than one-per-class. This issue is revisited in this paper where oneper-class class and pairwise coupling decomposition schemes used with both, SVMs and neural networks, are compared on a real life problem. Various methods for aggregating the results of pairwise classifiers are experimented. Beside our online handwriting application, experiments on some databases of the Irvine repository are also reported.