Title :
On the posterior-probability estimate of the error rate of nonparametric classification rules
Author :
Lugosi, Gábor ; Pawlak, Miroslaw
Author_Institution :
Dept. of Math., Budapest Tech. Univ., Hungary
fDate :
3/1/1994 12:00:00 AM
Abstract :
The posterior-probability estimate of the classification error rate of some nonparametric classification rules is studied. The variance of the estimator is shown to have same remarkable distribution-free properties for the k-nearest neighbor, kernel, and histogram rules. We also investigate the bias of the estimate and establish its consistency and upper bounds. The version of the estimate calculated from an independent set of unclassified patterns is also considered
Keywords :
error statistics; estimation theory; nonparametric statistics; pattern recognition; probability; classification error rate; distribution-free properties; estimate bias; histogram rules; k-nearest neighbor rule; kernel rules; nonparametric classification rules; posterior-probability estimate; unclassified patterns; upper bounds; variance; Error analysis; Histograms; Kernel; Mathematics; Pattern recognition; Random variables; Smoothing methods; Testing; Upper bound; Yield estimation;
Journal_Title :
Information Theory, IEEE Transactions on
