KNN is an implementation of the K-Nearest Neighbors algorithm. K-Nearest Neighbor is a method for classifying object based on training/tuning examples. Additionally, this implementation uses Forward Feature Selection for tuning optimal 'K' values based on training data.