GNU R racing for unbalanced methods selection

 A dataset is said to be unbalanced when the class of interest (minority
 class) is much rarer than normal behaviour (majority class). The cost of
 missing a minority class is typically much higher that missing a
 majority class. Most learning systems are not prepared to cope with
 unbalanced data and several techniques have been proposed. This package
 implements some of most well-known techniques and propose a racing
 algorithm to select adaptively the most appropriate strategy for a given
 unbalanced task.