Speeding up Support Vector Machines - Probabilistic versus Nearest Neighbour Methods for Condensing Training Data
Moïri Gamboni, Abhijai Garg, Oleg Grishin, Seung Man Oh, Francis Sowani, Anthony Spalvieri-Kruse, Godfried T. Toussaint, Lingliang Zhang
2014
Abstract
Several methods for reducing the running time of support vector machines (SVMs) are compared in terms of speed-up factor and classification accuracy using seven large real world datasets obtained from the UCI Machine Learning Repository. All the methods tested are based on reducing the size of the training data that is then fed to the SVM. Two probabilistic methods are investigated that run in linear time with respect to the size of the training data: blind random sampling and a new method for guided random sampling (Gaussian Condensing). These methods are compared with k-Nearest Neighbour methods for reducing the size of the training set and for smoothing the decision boundary. For all the datasets tested blind random sampling gave the best results for speeding up SVMs without significantly sacrificing classification accuracy.
DownloadPaper Citation
in Harvard Style
Gamboni M., Garg A., Grishin O., Man Oh S., Sowani F., Spalvieri-Kruse A., T. Toussaint G. and Zhang L. (2014). Speeding up Support Vector Machines - Probabilistic versus Nearest Neighbour Methods for Condensing Training Data . In Proceedings of the 3rd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-018-5, pages 364-371. DOI: 10.5220/0004927003640371
in Bibtex Style
@conference{icpram14,
author={Moïri Gamboni and Abhijai Garg and Oleg Grishin and Seung Man Oh and Francis Sowani and Anthony Spalvieri-Kruse and Godfried T. Toussaint and Lingliang Zhang},
title={Speeding up Support Vector Machines - Probabilistic versus Nearest Neighbour Methods for Condensing Training Data},
booktitle={Proceedings of the 3rd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2014},
pages={364-371},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004927003640371},
isbn={978-989-758-018-5},
}
in EndNote Style
TY - CONF
JO - Proceedings of the 3rd International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Speeding up Support Vector Machines - Probabilistic versus Nearest Neighbour Methods for Condensing Training Data
SN - 978-989-758-018-5
AU - Gamboni M.
AU - Garg A.
AU - Grishin O.
AU - Man Oh S.
AU - Sowani F.
AU - Spalvieri-Kruse A.
AU - T. Toussaint G.
AU - Zhang L.
PY - 2014
SP - 364
EP - 371
DO - 10.5220/0004927003640371