An Approach for Discretization and Feature Selection Of Continuous-Valued Attributes in Medical Images for Classification Learning

Abstract

Many supervised machine learning algorithms require a discrete feature space. In this paper, we review previous work on continuous feature discretization and, identify defining characteristics of the method. We then propose a new supervised approach which combines discretization and feature selection to select the most relevant features which can be used for classification purpose. The classification technique to be used is Associatve Classifiers. The features used are Harlick Texture features extracted from MRI Images. The results show that the proposed method is efficient and well-suited to perform preprocessing of continuous valued attributes.