Diagram
A 2x2 confusion matrix comparing ground-truth labels to model predictions, showing true positives, false negatives, false positives, true negatives, and row totals for labeled positives and negatives.
Labeled Data and Ground Truth
Labeled data is data where each item carries a known, correct outcome called ground truth. In a classification task, each item might be tagged as positive or negative. A confusion matrix compares model predictions against those ground-truth labels using four counts: true positives, true negatives, false positives, and false negatives. For any slice of data, the labeled negatives equal true negatives plus false positives, and the labeled positives equal false negatives plus true positives. Comparing confusion matrices between slices can reveal bias, such as better recall for one group than another.
Domain 3 of 3
36 practice questions
Practice this DomainLast updated