We assess the behaviour of 5 different feature extraction methods for an acoustic event classification task.built using the same SVM underlying technology - by means of two different techniques: accuracy and the entropy triangle. The entropy triangle is able to find a classifier instance whose relatively high accuracy stems from an attempt to specialize in some classes to the detriment of the overall behaviour. On all other cases, fair classifiers, accuracy and entropy triangle agree.
Cite as: Mejía-Navarrete, D., Gallardo-Antolín, A., Peláez-Moreno, C., Valverde-Albacete, F.J. (2011) Feature extraction assessment for an acoustic-event classification task using the entropy triangle. Proc. Interspeech 2011, 309-312, doi: 10.21437/Interspeech.2011-117
@inproceedings{mejianavarrete11_interspeech,
author={David Mejía-Navarrete and Ascensión Gallardo-Antolín and Carmen Peláez-Moreno and Francisco J. Valverde-Albacete},
title={{Feature extraction assessment for an acoustic-event classification task using the entropy triangle}},
year=2011,
booktitle={Proc. Interspeech 2011},
pages={309--312},
doi={10.21437/Interspeech.2011-117}
}