TY - JOUR
T1 - Linear discriminant feature extraction using weighted classification confusion information
AU - Lee, Hung Shin
AU - Chen, Berlin
PY - 2008
Y1 - 2008
N2 - Linear discriminant analysis (LDA) can be viewed as a two-stage procedure geometrically. The first stage conducts an orthogonal and whitening transformation of the variables. The second stage involves a principal component analysis (PCA) on the transformed class means, which is intended to maximize the class separability along the principal axes. In this paper, we demonstrate that the second stage does not necessarily guarantee better classification accuracy. Furthermore, we propose a generalization of LDA, weighted LDA (WLDA), by integrating the empirical classification confusion information between each class pair, such that the separability and the classification error rate can be taken into consideration simultaneously. WLDA can be efficiently solved by a lightweight eigen-decomposition and easily combined with other modifications to the LDA criterion. The experiment results show that WLDA can yield a relative character error reduction of 4.6% over LDA on the Mandarin LVCSR task.
AB - Linear discriminant analysis (LDA) can be viewed as a two-stage procedure geometrically. The first stage conducts an orthogonal and whitening transformation of the variables. The second stage involves a principal component analysis (PCA) on the transformed class means, which is intended to maximize the class separability along the principal axes. In this paper, we demonstrate that the second stage does not necessarily guarantee better classification accuracy. Furthermore, we propose a generalization of LDA, weighted LDA (WLDA), by integrating the empirical classification confusion information between each class pair, such that the separability and the classification error rate can be taken into consideration simultaneously. WLDA can be efficiently solved by a lightweight eigen-decomposition and easily combined with other modifications to the LDA criterion. The experiment results show that WLDA can yield a relative character error reduction of 4.6% over LDA on the Mandarin LVCSR task.
KW - Confusion information
KW - Feature extraction
KW - Linear discriminant analysis
KW - Speech recognition
UR - http://www.scopus.com/inward/record.url?scp=84867192980&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84867192980&partnerID=8YFLogxK
U2 - 10.21437/interspeech.2008-448
DO - 10.21437/interspeech.2008-448
M3 - Conference article
AN - SCOPUS:84867192980
SN - 2308-457X
SP - 2254
EP - 2257
JO - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
JF - Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
T2 - INTERSPEECH 2008 - 9th Annual Conference of the International Speech Communication Association
Y2 - 22 September 2008 through 26 September 2008
ER -