KDE-Based Ensemble Learning for Imbalanced Data

Imbalanced class distribution affects many applications in machine learning, including medical diagnostics, text classification, intrusion detection and others. In this paper, we propose a novel ensemble classification method designed to deal with imbalanced data. The proposed trains each tree the using uniquely generated synthetically balanced data balancing is carried out via kernel density estimation, which offers natural effective approach generating new sample points. We show that results lower variance of model estimator. tested against benchmark classifiers on range simulated real-life experiments classifier significantly outperforms methods.