Efficient and effective clustering algorithms by using sampling techniques for imbalanced data

Abstract

Data mining and Knowledge discovery is the process of discovering hidden knowledge from the datasets. In data mining the newlinemain approaches are supervised and unsupervised learning. In unsupervised learning k-means is one of the benchmark algorithms. newlineAmong the various real world data sources, many of the applications have the imbalance data and the imbalance data learning is one of the emerging fields of research which draws a special attention because of its significance. The present research have investigated through the area of this problem with a deeper insight newlineand found that there is still a great/ broad scope to propose novel and efficient unsupervised algorithms for better knowledge mining from data. The data mining research community is facing a burning challenge in Class Imbalance Learning (CIL) in unsupervised learning newlinedomain. The problem of class imbalance learning is of great significance when dealing with the real-world datasets. The data imbalance problem is more serious in the case of binary class, where the number of instances in one class predominantly outnumbers the number of instances in another class. Class imbalance learning is one of the issues that is to be addressed since it degrades the performance of several traditional unsupervised approaches. newline