Also known as: KNN Β· k Nearest Neighbors
KNN, or k-nearest neighbors, is a machine learning method that predicts an output for a new example based on the labels or values of its k most similar training examples. It is a lazy learning method because most computation occurs at prediction time rather than during a separate model-training phase.
A non-parametric supervised learning method first developed by Evelyn Fix and Joseph Hodges in 1951, and later expanded by Thomas Cover. It is used for classification and regression.