EoT: Ensemble of Trees For Classifying High Dimensional Imbalanced Data

Abstract

Class imbalance classification is a challenging research problem in data mining and machine learning, as most of the real-life datasets are often imbalanced in nature. Existing learning algorithms maximize the classification accuracy by correctly classifying the majority class, but misclassify the minority class. However, the minority class instances are representing the concept with more significant interest than the majority class instances in real-life applications. Recently, several techniques based on sampling methods (under-sampling of the majority class and over-sampling the minority class), cost-sensitive learning methods, and ensemble learning have been used in the literature for classifying imbalanced datasets. In this research, we introduce a new correlation-based feature grouping with under-sampling ensemble algorithm, called EoT for effective high dimensional imbalanced classification. EoT provides an competitive alternative to Bagging (sampling Bagging) and Random forest (sampling with Random forest) algorithms. We evaluated the performance of our proposed algorithm with the state-of-the-art methods based on ensemble like RUS-Bagging and RUS-Randomforest on 10 high dimensional imbalanced binary-class datasets with various imbalance ratios. The experimental results show that the EoT is a promising and effective approach for dealing with highly imbalanced datasets with large number of features.