Chapter 12 - An entropy-based hybrid feature selection approach for medical datasets

Abstract

Feature selection is an important step of data mining. The literature on feature selection techniques is very vast encompassing the applications of machine learning and pattern recognition. The techniques show that more information is not always good in machine learning applications. The ultimate goal of any feature selection method is to remove irrelevant or redundant features/attributes of datasets in order to reduce computation time and storage space, and to improve prediction performance of the learning algorithms. More generally, a feature selection algorithm is selected based on the following considerations—simplicity, stability, number of reduced features, classification accuracy, storage, and computational requirements.