Is it possible to use the Performance operator with the Compare ROCs one?
So a brief introduction to my question: I'm a computer science student and one of my classes is an introduction to data science and statistics. My first project is to use Rapid Miner to predict whether a given person will donate to a certain NGO. The professor gave us a sample data set with over 20k entries and a data set to use for the prediction. He gave us a couple instruction but said nothing much so were basically on our own. The objective is for us to try different prediction algorithms and then tell him which one we feel performs the best for that specific data set. We found out 90% of our time would be spent cleansing the data since it had a lot of inconsistencies. That wasn't a fun task. Now we feel like we have the data clean enough for our purpose but we're still lost, we don't know what criteria to use when we clean the data other than when the numbers are in the wrong format or there's irrelevant text on a certain field. Anyway, we found out a couple algorithms that would fit our plan quite well like Decision Tree, Naive Bayes, Random Fortest and Deep Learning and also that we could compare them using the Compare ROCs operator and analyze the ROC curve afterwards. Doing that, we figured out Decision Tree was probably the best one because of what we read about how to interpret ROC curves. We also began experimenting with the Performance operator and got a 95% performance out of the Decision Tree but we'd love to compare them all at the same time. My idea was the following: maybe I could use the ROCs operator that allows multiple prediction operators inside and apply them all the Performance operator at the same time so we could have the results all at once but I haven't quite figured out how. Is there any way to do what I want to do? Sorry for the long essay