Evaluation of Attribute Selection Methods with Tree based Supervised Classification-A Case Study with Mammogram Images

Abstract

Attribute selection is generally considered as a challenging work in the development of image data mining oriented applications. Attribute subset selection is mainly an optimization problem, which involves searching the space of possible feature subsets to select the one that is optimal or nearly optimal with respect to the performance measures accuracy, complexity etc., of the application. This paper presents a comparative evaluation of several attribute selection methods based on the performance accuracy of different tree based supervised classification for mammogram images of MIAS database