Title :
Comments on "The Extraction of Pattern Features from Imperfectly Identified Samples"
Author :
Toussaint, Godfried T.
Author_Institution :
School of Computer Science, McGill University
Abstract :
In a recent note1 C. C. Babu derives an upper bound on the probability of misclassification for the case when part of the training data is incorrectly labeled. Using this notation the bound is given by
Keywords :
Computer science; Control systems; Data mining; Distributed computing; Feature extraction; Gaussian distribution; Instruments; Optical computing; Training data; Upper bound;
Journal_Title :
Computers, IEEE Transactions on
DOI :
10.1109/T-C.1974.224069
