Title :
Suboptimum Linear Feature Selection in Multiclass Problem
Author :
Ichino, Manabu ; Hiramatsu, Keiji
Author_Institution :
Information Science Laboratory, Tokyo Electrical Engineering College, Tokyo, Japan.
Abstract :
A suboptimum method of linear feature selection in multiclass problem is presented. The set of features is selected in sequential manner based on an upper bound on the probability of error. The proposed method is applied to a problem of classifying Japanese vowels. Computer simulation results are presented and discussed.
Keywords :
Computer errors; Computer simulation; Covariance matrix; Density functional theory; Density measurement; Extraterrestrial measurements; Gaussian distribution; Pattern recognition; Upper bound; Vectors;
Journal_Title :
Systems, Man and Cybernetics, IEEE Transactions on
DOI :
10.1109/TSMC.1974.5408517
