Imbalance Data Classification Method Based on Cluster Boundary Sampling RF-Bagging
Peng Li, Tingting Bi, Jiuling Huang, Kaihui Zhang · 2014
This paper proposed a method based on the cluster boundary sampling RF-Bagging to solve the problem of imbalance data classification. It uses the cluster boundaries sampling of down-sampling preprocessing training data and then uses SVM and RF two different base classifiers as learning algorithms, integrated training and learning before and after sampling data by bagging respectively, two contrast experiment results are obtained. Finally, we use ROC curves and AUC values as results of the evaluation. The experimental results show that the method can improve the classification effect of classifier, and deal with the imbalance of data classification problems effectively.