Question: In dev-set error analysis, what does it mean for an example to be mislabeled, which part of the pair (x, y) is wrong, and give a concrete example.

Sample answer: An example is mislabeled when the human-assigned target is wrong before the model ever sees it. In the pair (x, y), the label y is incorrect, while x is the input data. For instance, a medical X-ray showing no fracture could be marked as "fracture present," or an X-ray with a clear fracture could be marked as "no fracture."

Key points:

• The mistake is introduced by the person who labeled the data.
• The incorrect part of (x, y) is y, the label.
• A positive/negative example can be flipped incorrectly, such as fracture vs. no-fracture.

Rubric: The answer should explain that mislabeling is a human labeling error that exists before the algorithm processes the data. It must identify y as the incorrect component of (x, y). It must also give a concrete mislabeling example.