An E-SMOTE technique for feature selection in High-Dimensional Imbalanced Dataset

Abstract

Feature Selection in High-Dimensional Imbalanced Dataset (where one class outnumbers the other class) plays an imperative task in field of Data mining and Bio-informatics. This paper proposes a new technique called E-SMOTE Technique for balancing the dataset and SVM classification for selecting the features. It is evaluated using micro array dataset.