Clustering Based Feature Selection Algorithm 1 Chandrashekhar W Gandhare, 2 Gaurav V Wanjare, 3 Mohit M Deshpande, 4 Ganesh V Padole

Abstract

-Feature selection is vital in the field of pattern classification due to accuracy and processing time considerations. The selection of proper features is of greater importance when the initial feature set is considerably large. Text classification is unsupervised machine learning method. It needs representation of objects and similarity measure, which compares distribution of features between objects. In this paper, we describe the hybrid method used for text clustering which is the combination of active feature selection, genetic algorithm and bisecting K-means. Internal quality measures compute the effectiveness of clustering. Our method is compared with K-means.