The Community for Technology Leaders
Green Image
ABSTRACT
<p><b>Abstract</b>—We derive a class of computationally inexpensive linear dimension reduction criteria by introducing a weighted variant of the well-known <it>K</it>-class <it>Fisher criterion</it> associated with <it>linear discriminant analysis</it> (LDA). It can be seen that LDA weights contributions of individual class pairs according to the Euclidian distance of the respective class means. We generalize upon LDA by introducing a different weighting function.</p>
INDEX TERMS
Linear dimension reduction, Fisher criterion, linear discriminant analysis, Bayes error, approximate pairwise accuracy criterion.
CITATION
R.p.w. Duin, R. Haeb-Umbach, Marco Loog, "Multiclass Linear Dimension Reduction by Weighted Pairwise Fisher Criteria", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 23, no. , pp. 762-766, July 2001, doi:10.1109/34.935849
90 ms
(Ver )