Combining Synthetic Minority Oversampling Technique and Subset Feature Selection Technique For Class Imbalance Problem

Abstract

Building an effective classification model when the high dimensional data is suffering from class imbalance problem is a major challenge. The problem is severe when negative samples have large percentages than positive samples. To surmount the class imbalance and high dimensionality issues in the dataset, we propose a SFS framework that comprises of SMOTE filters, which are used for balancing the datasets, as well as feature ranker for pre-processing of data. The framework is developed using R language and various R packages. Then the performance of SFS framework is evaluated and found that proposed framework outperforms than other state-of-the-art methods.