Re: k-nearest neighbor a supervised data mining method?

Rich F wrote:
> Sorry - I meant to post this to the group -

No problem. Has happened to me to.

> I guess what I mean is it combined as an algorithm with something like
> a k means clustering algorithm to establish the sets in the first
> place?

Ah, first you cluster a set of given items to establish the clusters and then you decide where to put new items. Undoubtedly somebody has already done this, but I dont' know of any concrete cases or when something like that could be useful.

• Jan Hidders