Also known as: One vs. all
One-vs.-all is a multiclass classification strategy that trains one binary classifier for each class. Each classifier distinguishes one target class from all other classes, and the combined outputs determine the final prediction.
Given a classification problem with N classes, a solution consisting of N separate binary classification model---one binary classification model for each possible outcome. For example, given a model that classifies examples as animal, vegetable, or mineral, a one-vs.-all solution would provide the following three separate binary classification models: - animal versus not animal - vegetable versus not vegetable - mineral versus not mineral