Fast training of support vector machines using sequential minimal optimization
Advances in kernel methods
A Tutorial on Support Vector Machines for Pattern Recognition
Data Mining and Knowledge Discovery
Journal of Global Optimization
Which is the best multiclass SVM method? an empirical study
MCS'05 Proceedings of the 6th international conference on Multiple Classifier Systems
Input space versus feature space in kernel-based methods
IEEE Transactions on Neural Networks
A comparison of methods for multiclass support vector machines
IEEE Transactions on Neural Networks
The pre-image problem in kernel methods
IEEE Transactions on Neural Networks
On-board analysis of uncalibrated data for a spacecraft at mars
Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining
Letters: Compact multi-class support vector machine
Neurocomputing
A New Variant of the Optimum-Path Forest Classifier
ISVC '08 Proceedings of the 4th International Symposium on Advances in Visual Computing
Multi-class Support Vector Machine Simplification
PRICAI '08 Proceedings of the 10th Pacific Rim International Conference on Artificial Intelligence: Trends in Artificial Intelligence
Reducing the run-time complexity of support vector data descriptions
IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
Progressive refinement for support vector machines
Data Mining and Knowledge Discovery
A discrete approach for supervised pattern recognition
IWCIA'08 Proceedings of the 12th international conference on Combinatorial image analysis
Ancient Chinese zither (guqin) music recovery with support vector machine
Journal on Computing and Cultural Heritage (JOCCH)
Fast support vector data descriptions for novelty detection
IEEE Transactions on Neural Networks
Engineering Applications of Artificial Intelligence
Fast Multiclass SVM Classification Using Decision Tree Based One-Against-All Method
Neural Processing Letters
Expert Systems with Applications: An International Journal
Condensed vector machines: learning fast machine for large data
IEEE Transactions on Neural Networks
A novel heuristic for building reduced-set SVMs using the self-organizing map
IWANN'11 Proceedings of the 11th international conference on Artificial neural networks conference on Advances in computational intelligence - Volume Part I
Fast automatic microstructural segmentation of ferrous alloy samples using optimum-path forest
CompIMAGE'10 Proceedings of the Second international conference on Computational Modeling of Objects Represented in Images
Fast opposite maps: an iterative SOM-Based method for building reduced-set SVMs
IDEAL'12 Proceedings of the 13th international conference on Intelligent Data Engineering and Automated Learning
Web-based closed-domain data extraction on online advertisements
Information Systems
On the optimization of multiclass support vector machines dedicated to speech recognition
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part II
Hi-index | 0.00 |
There are well-established methods for reducing the number of support vectors in a trained binary support vector machine, often with minimal impact on accuracy. We show how reduced-set methods can be applied to multiclass SVMs made up of several binary SVMs, with significantly better results than reducing each binary SVM independently. Our approach is based on Burges' approach that constructs each reduced-set vector as the pre-image of a vector in kernel space, but we extend this by recomputing the SVM weights and bias optimally using the original SVM objective function. This leads to greater accuracy for a binary reduced-set SVM, and also allows vectors to be "shared" between multiple binary SVMs for greater multiclass accuracy with fewer reduced-set vectors. We also propose computing pre-images using differential evolution, which we have found to be more robust than gradient descent alone. We show experimental results on a variety of problems and find that this new approach is consistently better than previous multiclass reduced-set methods, sometimes with a dramatic difference.