Changing threshold for classification

[arne_g99]

Hello there it’s me again,

I am trying to tweak my model in a way, that my random forest focuses on keeping the false negative number as low as possible since these have a much higer impact on my result than the false positives. Somewhere I read about a rule-node but I’m not sure how to apply it in a way, that my random forest prefers more FP in the result, if the FN is up to 50 times lower. Thank you very much in advance !

[ScottF]

Hi @arne_g99 -

You can use the Binary Classification Inspector to adjust the threshold and see how your confusion matrix (and associated metrics) change. Once you arrive at a reasonable threshold value, you can apply a threshold as you mention using the Rule Engine node.

Check out this discussion from @paolotamag and associated workflow:

Binary Classification Inspector predicting upon new data

Hi @ReidF, I am happy you are liking our new node and its view. The node is meant to give an interactive solution to find the best threshold and select a single model from its view. However it also compute a number a number of things regardless its view acting as some kind of a new “Scorer node” only for binary classification. In fact it outputs the settings for you to store and re-apply with other nodes (like for example a Rule Engine node) to a deployment model. First of all by default you …

Applying Optimized Threshold from Binary Calssification Inspector – paolotamag