What does probability calibration fix?

Question: What is the main goal of probability calibration? A) To increase model complexity B) To reduce the number of classes C) To make predicted probabilities better match observed outcome frequencies D) To remove all false positives Correct: C Explanation: A calibrated classifier gives probabilities that correspond to real-world frequencies. For example, among samples predicted with 0.8 probability, roughly 80% should belong to the positive class. Topic: advanced ML / calibration / classification