A systematic evaluation of univariate filter feature selection methods for leukemia datasets

The number of cases of blood cancer has increased in a scary way due to a failure in prediction. Filter feature selection is a crucial technique for the microarray gene expression analysis. Univariate filters are particularly appealing due to their simplicity and efficiency. This study aims to compare six univariate filters according to the number of selected features and the classification accuracy using three different types of classifiers: support vector machine, naive bayes and artificial neural network. This study used leukemia datasets in both binary and multi-class formats to investigate the performance of the studied filters.