Receiver operating characteristic

Receiver operating characteristic (ROC) is a way of measuring how well a machine learning algorithm can classify things. To explain what this means, let's say you have a machine learning algorithm that is trying to tell the difference between cats and dogs. ROC would measure how accurate the algorithm is at being able to tell the difference between cats and dogs. It would measure how often the algorithm gets it right, and how often the algorithm gets it wrong. This can be used to help improve and refine the algorithm, so that it can become better and better at telling the difference between cats and dogs.