Bayes Vector Quantizer (BVQ) is a machine learning algorithm specific for class-imbalance classifications. Classes that implement BVQ on this project are highly generalized and commented, so they can be easily used with any other library. On "ml/bvq/basic" you can find an example of a simple implementation.
This work was the subject of a university exam and it is based on the paper:
C. Diamantini and D. Potena, "Bayes Vector Quantizer for Class-Imbalance Problem," in IEEE Transactions on Knowledge and Data Engineering, vol. 21, no. 5, pp. 638-651, May 2009.
Please refer to it for the theory behind the algorithm.