2. Strategies for scaling up evolutionary instance reduction algorithms for data ,ining

Abstract

Evolutionary algorithms are adaptive methods based on natural evolution that may be used for search and optimization. As instance selection can be viewed as a search problem, it could be solved using evolutionary algorithms. In this chapter, we have carried out an empirical study of the performance of CHC as representative evolutionary algorithm model. This study includes a comparison between this algorithm and other non-evolutionary instance selection algorithms applied in different size data sets to evaluate the scaling up problem. The results show that the stratified evolutionary instance selection algorithms consistently outperform the non-evolutionary ones. The main advantages are: better instance reduction rates, higher classification accuracy and reduction in resources consumption.