An overview of classification algorithms for （分类算法的概述）.pdf

International Journal of Emerging Technology and Advanced Engineering Website: (ISSN 2250-2459, Volume 2, Issue 4, April 2012) An overview of classification algorithms for imbalanced datasets Vaishali Ganganwar Army Institute of Technology, Pune vaishaliloni@ Abstract— Unbalanced data set, a problem often found in The patient could lose his/her life because of the delay in real world application, can cause seriously negative effect the correct diagnosis and treatment. Similarly, if carrying a on classification performance of machine learning bomb is positive, then it is much more expensive to miss a algorithms. There have been many attempts at dealing with terrorist who carries a bomb to a flight than searching an classification of unbalanced data sets. In this paper we present innocent person. a brief review of existing solutions to the class-imbalance problem proposed both at the data and algorithmic levels. The unbalanced data set problem appears in many real Even though a common practice to handle the problem of world applications like text categorization, fault detection, imbalanced data is to rebalance them artificially by fraud detection, oil-spills detection in satellite images, oversampling and/or under-sampling, some researchers