The performance of our model, Chi-FRBCS-BigDataCS, has been tested in an experimental study including twenty-four imbalanced big data cases of study. These results corroborate the goodness of the integration of the approaches that are used to solve the imbalanced problem and big data separately, namely the usage of the MapReduce framework and cost-sensitive learning. Furthermore, the synergy between both strategies alleviates some data
intrinsic problems, like the small sample size problem, that are induced because of the way the learning is done.