The MDF discrimination measure: fisher in disguise

Authors:
Marco Loog;Robert P. W. Duin;Max A. Viergever
Affiliations:
Image Sciences Institute, University Medical Center Utrecht, HP E01.334, P.O. Box 85500, 3508 GA Utrecht, The Netherlands;Pattern Recognition Group, Faculty of Applied Sciences, Delft University of Technology, P.O. Box 5046, 2600 GA Delft, The Netherlands;Image Sciences Institute, University Medical Center Utrecht, HP E01.334, P.O. Box 85500, 3508 GA Utrecht, The Netherlands
Venue:
Neural Networks
Year:
2004

Citing 4
Cited 0

Introduction to statistical pattern recognition (2nd ed.)

Introduction to statistical pattern recognition (2nd ed.)
Multiclass Linear Dimension Reduction by Weighted Pairwise Fisher Criteria

IEEE Transactions on Pattern Analysis and Machine Intelligence
A closed-form neural network for discriminatory feature extraction from high-dimensional data

Neural Networks
Nonlinear feature extraction for pattern recognition applications

Nonlinear feature extraction for pattern recognition applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Recently, a discrimination measure for feature extraction for two-class data, called the maximum discriminating (MDF) measure (Talukder and Casasent [Neural Networks 14 (2001) 1201-1218]), was introduced.In the present paper, it is shown that the MDF discrimination measure produces exactly the same results as the classical Fisher criterion, on the condition that the two prior probabilities are chosen to be equal. The effect of unequal priors on the efficiency of the measures is also discussed.