Granular Computing in Decision Approximation

A striking feature of granular classifiers obtained by this approach is that preserving the accuracy of them on original data, they reduce substantially the size of the granulated data set as well as the set of granular decision rules.