0

*NOTE, this question has been asked before here, yet I wasn't any less confused after reading the answers. *

Is the k-nearest neighbour algorithm a discriminative or a generative classifier? My first thought on this was that it was generative, because it actually used Bayes's theorem to compute the posterior. Searching further for this, it seems like it is a discriminative model, but I couldn't find the explanation.

So is KNN discriminative first of all? And if it is, is that because it doesn't model the the priors or the likelihood?

101
  • 65
  • 4
  • This question is already answered here http://stats.stackexchange.com/questions/105979/is-knn-a-discriminative-learning-algorithm/106081#106081 – jpmuc Dec 14 '14 at 09:59
  • Cross-posted at http://datascience.stackexchange.com/questions/3702/k-nearest-neighbour. – whuber Dec 14 '14 at 22:55

0 Answers0