Comparative Analysis of Recent Data-Level Methods for Imbalance Classification

In recent days Machine Learning Technique has become one of the ubiquitous techniques used for solving different real-life problems. The Data sets used in Machine Learning techniques play a vital role in enhancing the efficiency and performance of the algorithm. An accurate outcome is not possible most of the time because of multiple factors. One of them may also be the nature of the data set. Real-world data sets are more cluttered, complex, and unorganized. Because of this reason, real-world data sets are imbalanced in nature. Machine learning experts believe that imbalanced data is a significant issue and researchers are working to tackle this. Three approaches are employed to fix this issue, those are Data level, Algorithm level, and Hybrid level approaches. We have made an effort to conduct a comparative study on recent data-level approaches and also try to conduct a thorough analysis of the undersampling and oversampling approaches.