DefinePK

Abstract

Feature selection has become the vital step in many data mining application for instances classification. Feature selection eliminates irrelevant attribute to obtain high quality features that may contribute in enhancing classification process and producing better classification results. This study is conducted with the intention to find out the most appropriate features that may lead to the best accuracy for various datasets of same domain, which is medical domain. During the experiments, comparisons were made between six benchmark feature selection methods based on eight medical datasets. Then, the performance were analyzed based on two machine learning algorithms; Naïve Bayes and KNN with and without feature selection in term of F-Measure and ROC on those medical datasets. From the experiment the optimum feature subsets are found. Moreover, the findings effectively support the fact that feature selection helps in increasing the classifier performance with existence of minimum number of features. However, no single feature selection methods that best satisfy all datasets and learning algorithms and this will simplify by assumption that features are independent for a given class variable. Hence, it still enables to obtain the optimal dimensionality of the feature subsets within the respective medical datasets.