Artificial Bee Colony Programming for Feature Selected Cancer Data Classification

Abstract

Feature selection provides model extraction by using the necessary / related features and improves classification. Feature selection is a desired process at eliminating irrelevant and redundant features in data. Classification is used to distribute the data in a balanced way among the various classes defined in the data as a result of the feature selection. In this paper, we investigated the feature selected classification performance in cancer data of recently proposed Artificial Bee Colony Programming and widely used Genetic Programming. The experimental results are compared with our previous work show that Artificial Bee Colony Programming have better performance Genetic Programming.

Keywords: Feature Selection; Classification; Genetic Programming; Artificial Bee Colony Programming.