Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

Thread Subject:
kmeans

Subject: kmeans

From: Ozge Taskan

Date: 31 May, 2010 08:48:03

Message: 1 of 2

Hi ,

I try to cluster my data which has dimensions (532,3). I have to use k means algorithm. In general, my problem is an assignment problem. I try to assign my clusters to 4 different modes. I started with k=4. So it is not difficult to look at each assignment possibility since it is just 4^4. Now I have to increase my cluster number, however my mode number will be 4 again. I thought that I need to know how I can find the optimal number of clusters. I could not find any algorithm to determine the optimal number of clusters. Is there any algorithm that I can use with Matlab?

thank you very much.

Ozge

Subject: kmeans

From: Oleg Komarov

Date: 31 May, 2010 09:57:03

Message: 2 of 2

"Ozge Taskan"
> Hi ,
>
> I try to cluster my data which has dimensions (532,3). I have to use k means algorithm. In general, my problem is an assignment problem. I try to assign my clusters to 4 different modes. I started with k=4. So it is not difficult to look at each assignment possibility since it is just 4^4. Now I have to increase my cluster number, however my mode number will be 4 again. I thought that I need to know how I can find the optimal number of clusters. I could not find any algorithm to determine the optimal number of clusters. Is there any algorithm that I can use with Matlab?
>
> thank you very much.
>
> Ozge

You're trying to find a solution that has no definite answer.
Check these links:
http://www.rd.dnc.ac.jp/~tunenori/doc/487-053.pdf
http://home.dei.polimi.it/matteucc/Clustering/tutorial_html/kmeans.html
http://www.csse.monash.edu.au/~roset/papers/cal99.pdf

Oleg

Tags for this Thread

What are tags?

A tag is like a keyword or category label associated with each thread. Tags make it easier for you to find threads of interest.

Anyone can tag a thread. Tags are public and visible to everyone.

Contact us