Application of Random-SMOTE on Imbalanced Data Mining
Jia Li, Hui Li, Jun-Ling Yu · 2011
The performance of many classifiers based on balanced data sets can't do well in imbalanced data sets. This article integrates the over-sampling method of Random-SMOTE (R-S), which is based on SMOTE method, in imbalanced data mining. We use the R-S method to increase the number of the minority randomly in the minority sample space until it is almost equal to the majority in data mining tasks. 5 UCI imbalanced data sets are balanced with the integrated data mining process. Log it algorithm is used for classification with these data sets. The result shows that the integrated use of R-S in data mining can improve the performance of the classifier significantly.