How should the company retrain the model to meet these requirements?
Set the target_recall hyperparameter to 90%. Set the binary_classifier_model_selection_criteria hyperparameter to recall_at_target_precision.
Set the target_precision hyperparameter to 90%. Set the binary_classifier_model_selection_criteria hyperparameter to precision_at_target_recall.
Use 90% of the historical data for training. Set the number of epochs to 20.
Set the normalize_label hyperparameter to true. Set the number of classes to 2.
Explanations:
Setting thetarget_recallhyperparameter to 90% ensures that at least 90% of the customers likely to purchase the new product are targeted. Thebinary_classifier_model_selection_criteriaset torecall_at_target_precisionhelps to optimize the model for recall while keeping precision at an acceptable level.
Setting thetarget_precisionto 90% and usingprecision_at_target_recalldoes not address the requirement to ensure 90% recall. The focus should be on recall, not precision, to meet the campaign’s targeting goals.
Using 90% of the historical data for training and setting the number of epochs to 20 does not directly impact recall. While more data and training might improve model performance, the focus should be on adjusting recall, not on training data size or epochs.
Settingnormalize_labelto true and specifying 2 classes is unrelated to adjusting the model for the required recall rate. This option does not focus on optimizing recall and precision, which is crucial to meeting the campaign’s requirements.
I suspect that the answer is:
Set the target_recall hyperparameter to 90%. Set the binary_classifier_model_selection_criteria hyperparameter to recall_at_target_precision.