Class Imbalance Problem in Data Mining: Review

In last few years there are major changes and evolution has beendone on classification of data. As the application area oftechnology is increases the size of data also increases.Classification of data becomes difficult because of unboundedsize and imbalance nature of data. Class imbalance problembecome greatest issue in data mining. Imbalance problem occurwhere one of the two classes having more sample than otherclasses. The most of algorithm are more focusing onclassification of major sample while ignoring or misclassifyingminority sample. The minority samples are those that rarelyoccur but very important. There are different methods availablefor classification of imbalance data set which is divided intothree main categories, the algorithmic approach, datapreprocessingapproach and feature selection approach. Each ofthis technique has their own advantages and disadvantages. Inthis paper systematic study of each approach is define whichgives the right direction for research in class imbalance problem.