Selection of Classifier and Feature Selection Method for Microarray Data

Micro array data have a low instance-count and high dimensionality problem which prevent classifiers from building accurate models. This may result in significantly different classification accuracies across classifiers and features chosen. Therefore it is important to select the classifier and feature selection method that perform well on a specific data set. This paper proposes a novel criterion to select the best classifier and feature selection method for a specific micro array data set. Also a novel voting strategy to use the proposed selection criterion is presented. The experimental results show that the proposed criterion and voting method substantially increase classification accuracies for micro array data sets in the experiments.