Precision is a performance metric for classification models. It measures the accuracy of the positive predictions. Specifically, it is the ratio of true positive predictions to the total number of positive predictions (True Positives + False Positives). It answers the question: 'Of all the instances the model claimed were positive, how many actually were?'
Originated in the field of Information Retrieval (Kent et al., 1955) to evaluate the relevance of search results.
Precision is a standard metric in machine learning, especially important in scenarios where false positives are costly, such as email spam detection (you don't want to mark a legitimate email as spam).