[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Data clustering
From: |
Juan Pablo Carbajal |
Subject: |
Re: Data clustering |
Date: |
Sat, 21 Jun 2014 14:40:59 +0200 |
On Sat, Jun 21, 2014 at 12:17 PM, CdeMills <address@hidden> wrote:
> Hello,
>
> I would like to have some hints about the following problem. I test an
> electrical system with a number of input levels. The system response consist
> mainly of output steps, separated by short transition zones.
>
> As the command and the recording are asynchronous, I would like to divide
> the output data into zones where we are observing constant input level. The
> hypothesis are:
> 1) the number of input steps is known;
> 2) the number of points in the stable zones is greater than the number of
> points in the transition regions.
>
> Is there some algorithm in Octave permitting to compute such a map between
> points and level ? Basically the problem dimension is one.
>
> Regards
>
> Pascal
>
>
>
> --
> View this message in context:
> http://octave.1599824.n4.nabble.com/Data-clustering-tp4664909.html
> Sent from the Octave - General mailing list archive at Nabble.com.
>
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/help-octave
I hva edone similr stuff in very different ways. For examlpe a direct
use of diff can do it or running averages of the L2 norm derivative on
the smoothed output, even moving histrograms can help you decide when
you are stable. It all depends a lot on the time scale.
If you want to go data driven, you can try kmeans and then detect the
outliers from each cluster (they probably will be transitional
values). Try also the soft partition fcm in the fuzzy logic package.
Can you send a example data or plot?
- Data clustering, CdeMills, 2014/06/21
- Re: Data clustering,
Juan Pablo Carbajal <=