Confusion Matrix Based Entropy Correction in Multi-stream Combination

An MLP classifier outputs a posterior probability for each class. With noisy data, classification becomes less certain, and the entropy of the posteriors distribution tends to increase providing a measure of classification confidence. However, at high noise levels, entropy can give a misleading indication of classification certainty. Very noisy data vectors may be classified systematically into classes which happen to be most noise-like and the resulting confusion matrix shows a dense column for each noise-like class. In this article we show how this pattern of misclassification in the confusion matrix can be used to derive a linear correction to the MLP posteriors estimate. We test the ability of this correction to reduce the problem of misleading confidence estimates and to enhance the performance of entropy based full-combination multi-stream approach. Better word-error-rates are achieved for Numbers95 database at different levels of added noise. The correction performs significantly better at high SNRs.

Published in:
Proceedings of ISCA European Conference on Speech Communication and Technology (Eurospeech)
Presented at:
Proceedings of ISCA European Conference on Speech Communication and Technology (Eurospeech)
Geneve, Switzerland
IDIAP-RR 2002 53

Note: The status of this file is: Anyone

 Record created 2006-03-10, last modified 2020-10-28

Download fulltextPDF
External links:
Download fulltextURL
Download fulltextRelated documents
Rate this document:

Rate this document:
(Not yet reviewed)