The K Nearest Neighbors platform predicts a response value for a given observation using the responses of the observations in that observation’s local neighborhood. It can be used with a categorical response for classification and with a continuous response for prediction.

K Nearest Neighbors is a nonparametric method that is based on the distance to neighboring observations. Because of this fact, K Nearest Neighbors is able to classify observations using irregular predictor value boundaries. However, the algorithm is sensitive to irrelevant predictors, so selection of predictors can be beneficial before implementing K Nearest Neighbors.

K Nearest Neighbors has been used successfully in many applications, such as classifying satellite imagery and EKG patterns.