Simple incremental instance selection wrapper for classification

Abstract

Instance selection methods are very useful data mining tools for dealing with large data sets. There exist many instance selection algorithms capable for significant reduction of training data size for particular classifier without generalization degradation. In opposition to those methods, this paper focuses on general pruning methods which can be successfully applied for arbitrary classification method. Simple but efficient wrapper method based on generalization of Hart's Condensed Nearest Neighbors rule is presented and impact of this method on classification quality is reported. © 2012 Springer-Verlag Berlin Heidelberg.