A method for solving the class imbalance Problem in Classification Techniques

Class imbalance learning refers to learning from imbalanced data sets, in which some classes of examples (minority) are highly under-represented comparing to other classes (majority). The Very skewed class distribution degrades the learning ability of many traditional machine learning methods, especially in the recognition of examples from the minority classes, which are often deemed to be more important and interesting. Although quite a few ensemble learning approaches have been proposed to handle the problem, no in-depth research exists to explain why and when they can be helpful. We investigate mathematical links between single-class performance and ensemble diversity. One method to tackle this problem consists to resample the original training set, either by over-sampling the minority class and/or under-sampling the majority class. we propose two ensemble models (using a modular neural network and the nearest neighbor rule) trained on datasets under-sampled with genetic algorithms. Experiments with real datasets demonstrate the effectiveness of the methodology here proposed.