Instance Selection Using Evolutionary Algorithms: An Experimental Study

TitleInstance Selection Using Evolutionary Algorithms: An Experimental Study
Publication TypeBook Chapter
Year of Publication2005
AuthorsCano, J. R., Herrera F., and Lozano Manuel
EditorPal, Nikhil R., and Jain Lakhmi
Book TitleAdvanced Techniques in Knowledge Discovery and Data Mining
Pagination127–152
PublisherSpringer London
CityLondon
ISBN Number978-1-84628-183-9
Abstract

In this chapter, we carry out an empirical study of the performance of four representative evolutionary algorithm models considering two instance-selection perspectives, the prototype selection and the training set selection for data reduction in knowledge discovery. This study includes a comparison between these algorithms and other nonevolutionary instance-selection algorithms. The results show that the evolutionary instance-selection algorithms consistently outperform the nonevolutionary ones, offering two main advantages simultaneously, better instance-reduction rates and higher classification accuracy.

URLhttps://doi.org/10.1007/1-84628-183-0_5
DOI10.1007/1-84628-183-0_5