AWSMOTE: An SVM-Based Adaptive Weighted SMOTE for Class-Imbalance Learning

In class-imbalance learning, Synthetic Minority Oversampling Technique (SMOTE) is a widely used technique to tackle problems from the data level, whereas SMOTE blindly selects neighboring minority class points when performing an interpolation among them and inevitably brings collinearity between generated new original ones. To combat these problems, we propose in this study adaptive-weighting method, termed as AWSMOTE. AWSMOTE applies two types of SVM-based weights into SMOTE. A kind weight variable space drawbacks collinearity, while another utilized sample purposefully choose those support vectors interpolation. compared with its improved versions six simulated datasets 22 real-world datasets. The results demonstrate effectiveness advantages proposed approach.