Abstract
A new linear dimensionality reduction (LDR) technique for pattern classification and machine learning is presented, which, though linear, aims at maximizing the Chernoff distance in the transformed space. The corresponding two-class criterion, which is maximized via a gradient-based algorithm, is presented and initialization procedures are also discussed. Empirical results of this and traditional LDR approaches combined with two well-known classifiers, linear and quadratic, on synthetic and real-life data show that the proposed criterion outperforms the traditional schemes.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Aladjem, M.: Linear Discriminant Analysis for Two Classes Via Removal of Classification Structure. IEEE Trans. on Pattern Analysis and Machine Intelligence 19(2), 187–192 (1997)
Chong, E., Zak, S.: An Introduction to Optimization, 2nd edn. John Wiley and Sons, Inc., New York (2001)
Cooke, T.: Two Variations on Fisher’s Linear Discriminant for Pattern Recognition. IEEE Transations on Pattern Analysis and Machine Intelligence 24(2), 268–273 (2002)
Du, Q., Chang, C.: A Linear Constrained Distance-based Discriminant Analysis for Hyperspectral Image Classification. Pattern Recognition 34(2), 361–373 (2001)
Duda, R., Hart, P., Stork, D.: Pattern Classification, 2nd edn. John Wiley and Sons, Inc., New York (2000)
Gao, H., Davis, J.: Why Direct LDA is not Equivalent to LDA. Pattern Recognition 39, 1002–1006 (2006)
Herrera, M., Leiva, R.: Generalización de la Distancia de Mahalanobis para el Análisis Discriminante Lineal en Poblaciones con Matrices de Covarianza Desiguales. Revista de la Sociedad Argentina de Estadística 3(1-2), 64–86 (1999)
Loog, M., Duin, P.W.: Linear Dimensionality Reduction via a Heteroscedastic Extension of LDA: The Chernoff Criterion. IEEE Transactions on Pattern Analysis and Machine Intelligence 26(6), 732–739 (2004)
Lotlikar, R., Kothari, R.: Adaptive Linear Dimensionality Reduction for Classification. Pattern Recognition 33(2), 185–194 (2000)
Newman, D., Hettich, S., Blake, C., Merz, C.: UCI repository of machine learning databases, University of California, Irvine, Dept. of Computer Science (1998)
Rao, A., Miller, D., Rose, K., Gersho, A.: A Deterministic Annealing Approach for Parsimonious Design of Piecewise Regression Models. IEEE Transactions on Pattern Analysis and Machine Intelligence 21(2), 159–173 (1999)
Raudys, S.: Evolution and Generalization of a Single Neurone: I. Single-layer Perception as Seven Statistical Classifiers. Neural Networks 11(2), 283–296 (1998)
Rueda, L.: Selecting the Best Hyperplane in the Framework of Optimal Pairwise Linear Classifiers. Pattern Recognition Letters 25(2), 49–62 (2004)
Rueda, L., Herrera, M.: Linear Discriminant Analysis by Maximizing the Chernoff Distance in the Transformed Space. Submitted for Publication (2006), Electronically available at http://www.inf.udec.cl/~lrueda/papers/ChernoffLDAJnl.pdf
Rueda, L., Oommen, B.J.: On Optimal Pairwise Linear Classifiers for Normal Distributions: The Two-Dimensional Case. IEEE Transations on Pattern Analysis and Machine Intelligence 24(2), 274–280 (2002)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Rueda, L., Herrera, M. (2006). A New Linear Dimensionality Reduction Technique Based on Chernoff Distance. In: Sichman, J.S., Coelho, H., Rezende, S.O. (eds) Advances in Artificial Intelligence - IBERAMIA-SBIA 2006. IBERAMIA SBIA 2006 2006. Lecture Notes in Computer Science(), vol 4140. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11874850_34
Download citation
DOI: https://doi.org/10.1007/11874850_34
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-45462-5
Online ISBN: 978-3-540-45464-9
eBook Packages: Computer ScienceComputer Science (R0)