6
$\begingroup$

I have 100 data points, observed on 15 variables. I want to cluster my 100 observations, but I am unable to visualise 15-dimensional clusters in MATLAB.

$\endgroup$
  • $\begingroup$ There is nothing specific to clustering in this question... $\endgroup$ – Has QUIT--Anony-Mousse Sep 23 '15 at 15:09
  • 2
    $\begingroup$ 3 upvotes for this "well researched" question? Really? $\endgroup$ – Alec Teal Sep 23 '15 at 22:41
12
$\begingroup$

Calculate distances between data points, as appropriate to your problem.

Then plot your data points in two dimensions instead of fifteen, preserving distances as far as possible. This is probably the key aspect of your question. Read up on multidimensional scaling (MDS) for this.

Finally, color your points according to cluster membership.

| cite | improve this answer | |
$\endgroup$
2
$\begingroup$

I have successfully used a Self-Organizing Map (SOM) in the past for this task. It is a kind of Neural Network with some relation to Clustering, with significant advantages over them for some specific tasks. The main advantage (to me) is that it is an unsupervised method, meaning that you can apply it even with unknown classes in your data. If you know your classes, you could use this information to color the distinctive clusters/regions obtained in the output map.

https://en.wikipedia.org/wiki/Self-organizing_map

| cite | improve this answer | |
$\endgroup$
  • 1
    $\begingroup$ Welcome. It is a pleasure to be helpful. If it has been useful to you, please, mark it as solved. Thanks. $\endgroup$ – shirowww Sep 24 '15 at 15:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.