A method for the linear discrimination of two classes is presented. It maximizes the Patrick-Fisher (PF) distance between the projected class-conditional densities. Since the PF distance is a highly nonlinear function, we propose a method, which searches for the directions corresponding to several large local maxima of the PF distance. Its novelty lies in a neural network transformation of the data along a found direction into data with deflated maxima of the PF distance and iteration to obtain the next direction. A simulation study indicates that the method has the potential to find the global maximum of the PF distance.
CITATION STYLE
Aladjem, M. (1998). Linear discriminant analysis for two classes via recursive neural network reduction of the class separation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 1451, pp. 775–784). Springer Verlag. https://doi.org/10.1007/bfb0033302
Mendeley helps you to discover research relevant for your work.