hi,
for thin datasets, I have heard from data augmentation, e.g. creating data from existing data, It is applied with statistical techniques to create new data base on mean, variance, etc. of the attributes from the original data.
does there exist a similar operator to create augmented data based on some example set in Rapidminer?
http://www.stat.harvard.edu/Faculty_Content/meng/JCGS01.pdf