In the context of model evaluation, what does 'calibration' refer to?
Select an answer to reveal the explanation.
Short Explanation
Here's the deal — calibration is about reliability. If you predict 70% across many cases, about 70% should actually happen, so the predicted-probabilities-match-observed-frequencies option is right.
Full Explanation
Calibration checks whether predicted probabilities are reliable: if you predict 70% across many cases, about 70% should occur. Ranking positives above negatives is discrimination, often AUC. Tuning hyperparameters on a validation set is model selection. Standardizing inputs to mean 0 and standard deviation 1 is feature normalization.