CLASS IMBALANCE PROBLEM IN DATA SCIENCE: REVIEW

In last few years there are many changes and evolution has been done on the classification of data. Application area of technology increases then the size of data also increases. Classification of data becomes difficult because of imbalance nature and unbounded size of data. Class imbalance problem becomes the greatest issue in the data mining. Imbalance problem occurs where one of the two classes having more sample than the other classes. The most of the algorithms are focusing on the classification of major sample while ignoring minority sample. The minority samples are those samples that rarely occur but are very important. There are different methods available for the classification of imbalance data set which is divided into three categories i.e. the algorithmic approach, feature selection approach and the data processing approach. These approaches have their own advantages and disadvantages. In this paper a systematic study of each process is defined which gives the right direction for research in class imbalance problem.