Sankar Vema - Blog Space

As discussed in my previous blog, Imbalanced data poses serious challenges in Machine Learning .  One of approach to combat this imbalance is data is to alter the training set in such a way as to create a more balanced class distribution so that the resulting sampled data set can be used with traditional data-mining algorithms. This can be achieved through...  Under-sample where the size of the majority class is reduced using different techniques like reducing redundancy, removing boundary candidates etc., Over-sample where the size of the minority class is increased by adding more candidates which can augment the data set. Hybrid approach where a combination of both oversampling of minority class and under sampling of majority class is attempted. Each of these techniques discussed below Random Over Sampling In random over-sampling, the minority class instances are duplicated in the data set until a more balanced distribution is reached. As a illustration, consider...

For many real world machine learning problem we see an imbalance in the data where one class under represented in relative to others. This leads to mis-classification of elements between classes. The cost of mis-classification is often unknown at learning time and can be far too high. We often see this type of imbalanced classification scenarios in fraud/intrusion detection, medical diagnosis/monitoring, bio-informatics, text categorization and et al. To better understand the problem, consider the “Mammography Data Set,” a collection of images acquired from a series of mammography examinations performed on a set of distinct patients. For such a data set, the natural classes that arise are “Positive” or “Negative” for an image representative of a “cancerous” or “healthy” patient, respectively. From experience, one would expect the number of noncancerous patients to exceed greatly the number of cancerous patients; indeed, this data set contains 10,923 “Negative” (majority class) and...