Gustavo A Valencia-Zapata, Daniel Mejia, Gerhard Klimeck, Michael Zentner, Okan Ersoy

Probabilistic mixture models have been widely used for different machinelearning and pattern recognition tasks such as clustering, dimensionalityreduction, and classification. In this paper, we focus on trying to solve themost common challenges related to supervised learning algorithms by usingmixture probability distribution functions. With this modeling strategy, weidentify sub-labels and generate synthetic data in order to reach betterclassification accuracy. It means we focus on increasing the training datasynthetically to increase the classification accuracy.