How is KNN different from k-means clustering?
Options
Answers
-
While k-means clustering is an unsupervised clustering approach, k-nearest neighbors are a supervised classification algorithm. Even though the methods may initially appear to be comparable, this means that for K-Nearest Neighbors to function, you must have labelled data that you intend to use to categorize an unlabeled point (thus the nearest neighbors component). K-means clustering just needs a collection of unlabeled points and a threshold; it then uses the mean of the distances between the points to gradually learn how to sort the unlabeled points into groups.