Comparing Nodes Precision and recall


Many time when we are working about Dataset we are checking 10-20 different algorithms  from Wekathan we use to take the Threshold analyzer of the Knime and to check with each threshold can we get best recall , while we want to get to precision of 0.7-0.85 , we are checking manually one by one which algo' is the best andI would like to know if there is a way to create automatic  Comparison  between the algorithm nodes and we pick best one of them for there a node that make a  Comparison and pick the best  algorithms  ?


thank you


have a look at the example Workflow for the PMML webinar. Among other things, it demonstrates how PMML ensembles could be used for exactly this purpose.

can you please be more specific where can I learn how to do that on this webinar?

I was looking on this video , are talking about this video?

Well it's from the same webinar I meant, but I was more specifically referring to the corresponding example workflow and especially the Model Combination (Ensembles) section of that. If that isn't enough for you, the search terms pmml and ensemble together might be able to help.