This workflow creates a model selction on the webportal, where data analysts can choose a training and a test set between the datasets of 2006 to 2012, the target variable and the algorithms, which should be considered for the model comparison. The results are then displayed in a ROC Curve. For further comparison two algorithms can be chosen to compare them via a Lift Chart, before the data analysts can decide for a final algorithm.
This is a companion discussion topic for the original entry at https://kni.me/w/pGsLrYoxE1AVyS7I