Positive Class

noun

say it: “POZ-ih-tiv KLAS” /ˈpɑzətɪv ˈklæs/

In a classification with two answers, the category a model is meant to detect, such as fraud or disease, usually labeled 1.

What is the positive class?

The positive class is, in a two-answer classifier, the answer you're looking for, such as fraud or spam, usually labeled 1. "Positive" doesn't mean good: it means found. Precision, recall and the confusion matrix's true and false positives are all about it.

In a sentence

  • “In the spam filter, spam is the positive class and real mail is the negative one.”
  • “Fraud was the positive class, so a true positive meant a fraud the check caught.”

Level: Journeyman · More AI Basics terms at this level

Updated