Asked about Where ground truth comes from

How do you write labelling guidelines?

An answer that works

Define each label with concrete positive and negative examples rather than adjectives, state explicitly how to handle ambiguity, include a not-applicable option, then have someone else label with them and fix whatever they got wrong.

Reading an answer is not being able to give one. The section that teaches this is Where ground truth comes from, 5 hours of Data quality & labelling.

Learn this properly

Also asked about Data quality & labelling

Two annotators disagree on 30 percent of labels. What does that tell you?You suspect many labels in your training set are wrong. How do you find them?How would you check for train-test contamination?