The positive class is the target class or condition that a classifier is designed to detect. The term does not necessarily mean a desirable outcome; in medical or security contexts, the positive class may represent a harmful event.
The class you are testing for. For example, the positive class in a cancer model might be "tumor." The positive class in an email classification model might be "spam." Contrast with negative class. The term positive class can be confusing because the "positive" outcome of many tests is often an undesirable result. For example, the positive class in many medical tests corresponds to tumors or diseases. In general, you want a doctor to tell you, "Congratulations! Your test results were negative." Regardless, the positive class is the event that the test is seeking to find. Admittedly, you're simultaneously testing for both the positive and negative classes.