A New Contrast Pattern-Based Classification for Imbalanced Data
Xiangtao Chen, Yajing Gao, Siqi Ren · 2018
Contrast pattern-based classifiers become more understandable and accurate on binary classification. However, these classifiers do not achieve good performance on class imbalance problems. Thus, this paper introduces a new contrast pattern-based classifier for class imbalance problems. The proposed method selects the appropriate contrast patterns by quality measures. Then we combine the quality measure of the pattern and class confidence proportion with the class imbalance level at the classification stage of the model. The simulation results show that our proposed outperforms the current contrast pattern-based classifiers and other state-of-the-art classifiers not directly based on contrast patterns for class imbalance problems.