Alleviating class imbalance problem in data mining

The class imbalance problem in two-class data sets is one of the most important problems. When samples of one class in a training data set vastly outnumber samples of the other class, standard machine learning algorithms tend to be overwhelmed by the majority class and ignore the minority class. There are several algorithms to alleviate the problem of class imbalance in literature. In this paper experiments have been done comparing the existing algorithms with each other and the algorithm which has the best performance tried to be found.