Synopsis
Finds a threshold for given prediction confidences (soft predictions) , costs and distributional information in order to turn it into a crisp classification. The optimization step is based on ROC analysis.
Description
This operator finds the best threshold for crisp classifying based on user defined costs.
Input
- example set: expects: ExampleSet, expects: ExampleSet
Output
- example set:
- threshold:
Parameters
- misclassification costs first: The costs assigned when an example of the first class is classified as one of the second.
- misclassification costs second: The costs assigned when an example of the second class is classified as one of the first.
- show roc plot: Display a plot of the ROC curve.
- use example weights: Indicates if example weights should be used.
- roc bias: Determines how the ROC (and AUC) are evaluated: Count correct predictions first, last, or alternatingly