A new sampling technique and SVM classification for feature selection in high-dimensional Imbalanced dataset

Feature selection in high-dimensional Imbalanced dataset (where one class highly outnumbers the other class) is an exigent task in data mining. Feature selection refers to selecting a subset of features from the original dataset. This paper focus on two problems i) Balancing the dataset ii) extracting the features. A new technique called Evolutionary sampling technique [EST] is developed to balance the dataset and Support Vector Machine [SVM] classification is used to calculate the accuracy and also to overcome the over fitting problem while sampling the dataset. The techniques are evaluated on a micro array dataset.