I have gone through many of your tutorial in the course of my research. I say they are Bravos and easy to learn.! Thanks to all for the job well-done!!!
However, I need your advice on how I can supply independent training and test sets as two data sources in the Weka KnowledgeFlow Environment to compare my AUC from 5 classifiers. I want a process that would train the classifiers through trainset and evaluate the performance on an independent testset (which has to be supplied by the user) and thereafter obtain the AUCs of the results in one chart (for ease of comparison) and legend indicated on one side. Please see the figure below from R.
The tutorial I followed was able to guide me through cross-validation process without user supplied test set.