Definition
The proportion of positive predictions made by a classification model that were actually correct (True Positives), measuring the model's exactness.
In short: The proportion of positive predictions made by a classification model that were actually correct (True Positives), measuring the model's exactness.