Hi, I have a question of classification on imbalanced dataset. I am wondering if there is a package which can solve this problem via sampling approach, like one-sided selection. A follow-up question is, how to select those 'representative' samples and remove noise/borderlines and redundancy in order to increase classification accuracy. Is there any work which has been implemented in R or some GNU softwares? Thanks, weiwei -- Weiwei Shi, Ph.D "Did you always know?" "No, I did not. But I believed..." ---Matrix III