The Community for Technology Leaders
RSS Icon
Subscribe
Issue No.07 - July (2002 vol.24)
pp: 893-904
ABSTRACT
<p>We consider a popular approach to multicategory classification tasks: a two-stage system based on a first (global) classifier with rejection followed by a (local) nearest-neighbor classifier. Patterns which are not rejected by the first classifier are classified according to its output. Rejected patterns are passed to the nearest-neighbor classifier together with the {\rm{top}}\hbox{-}h ranking classes returned by the first classifier. The nearest-neighbor classifier, looking at patterns in the {\rm{top}}\hbox{-}h classes, classifies the rejected pattern. An editing strategy for the nearest-neighbor reference database, controlled by the first classifier, is also considered. We analyze this system, showing that even if the first level and nearest-neighbor classifiers are not optimal in a Bayes sense, the system as a whole may be optimal. Moreover, we formally relate the response time of the system to the rejection rate of the first classifier and to the other system parameters. The error-response time trade-off is also discussed. Finally, we experimentally study two instances of the system applied to the recognition of handwritten digits. In one system, the first classifier is a fuzzy basis functions network, while in the second system it is a feed-forward neural network. Classification results as well as response times for different settings of the system parameters are reported for both systems.</p>
INDEX TERMS
Multicategory classification, rejection, global and local classification, hierarchical classifier, Bayes classifier.
CITATION
Nicola Giusti, Francesco Masulli, Alessandro Sperduti, "Theoretical and Experimental Analysis of a Two-Stage System for Classification", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol.24, no. 7, pp. 893-904, July 2002, doi:10.1109/TPAMI.2002.1017617
5 ms
(Ver 2.0)

Marketing Automation Platform Marketing Automation Tool