Quantcast
Channel: Machine Learning
Viewing all articles
Browse latest Browse all 62750

Question on Clustering with unknown clusters.

$
0
0

Hey guys, quick question. I have data that (in frequency) looks like : http://imgur.com/a89Sd

Upon visual inspection of the data I used K-Means clustering to separate peaks around A and C.

Is there such an algorithm that will detect the number of means to separate, ignoring false peaks B and D.

The final data has many more peaks than the sample data.

submitted by cyborgbrain
[link] [12 comments]

Viewing all articles
Browse latest Browse all 62750

Trending Articles