Sciweavers

ANNPR
2008
Springer

Feature Ranking Ensembles for Facial Action Unit Classification

13 years 5 months ago
Feature Ranking Ensembles for Facial Action Unit Classification
Recursive Feature Elimination RFE combined with feature-ranking is an effective technique for eliminating irrelevant features. In this paper, an ensemble of MLP base classifiers with feature-ranking based on the magnitude of MLP weights is proposed. This approach is compared experimentally with other popular feature-ranking methods, and with a Support Vector Classifier SVC. Experimental results on natural benchmark data and on a problem in facial action unit classification demonstrate that the MLP ensemble is relatively insensitive to the feature-ranking method, and simple ranking methods perform as well as more sophisticated schemes. The results are interpreted with the assistance of bias/variance of 0/1 loss function.
Terry Windeatt, Kaushala Dias
Added 12 Oct 2010
Updated 12 Oct 2010
Type Conference
Year 2008
Where ANNPR
Authors Terry Windeatt, Kaushala Dias
Comments (0)