Dear Ted,
Thank you very much for your answer. It is very inspiring for a
beginner like me to see the effort that you put to answer to questions
like mine, I'm sure that they look trivial. And the whole community
involved is great.
So, to summarize, my idea of K-medoids with DTW as a
Anand,
That is a fine idea. It is called a medoid instead of a mean (
https://en.wikipedia.org/wiki/Medoid )
The basic idea is that for any metric m, you can define the medoid as the
element from the set that minimizes the sum of the distances to the other
elements for that metric. In
Ted,
Thank you for your answer. Maybe I made a wrong picture about my data
when giving sinusoid as an example, my time series are not periodic.
Let's say that I have a signal that represents value of power when some
device is turned on. That power signal depends of the time when person
turns
Perhaps you could think of the centroid as one of the signals itself, from
which the average distance to rest of the signals in the cluster is the
lowest? This way instead of finding that mythical mean of DTWs, you hop
from one signal to another over iterations as you refine memberships.
However
Ted,
It's because I'm trying to cluster time series that may differ in
length, some parts may be shifted, some parts may last longer in one
signal than in the other (somehow skewed), but they represent more-less
the same thing. DTW seems good because it's optimized for such things
(used a
On Sat, Jan 10, 2015 at 3:02 AM, Marko Dinic marko.di...@nissatech.com
wrote:
For example, mean of two sinusoids while one of them is shifted by Pi is
0. And that's definitely not a good centroid in my case.
Well, if you think that phase shifts represent small distance proportional
to phase
Why is it you can't compute a mean?
On Fri, Jan 9, 2015 at 5:03 AM, Marko Dinic marko.di...@nissatech.com
wrote:
Thank you for your answer Ted.
What about some kind of Bisecting k-means? I'm trying to cluster time
series of different length and I came up to an idea to use DTW as a
Thank you for your answer Ted.
What about some kind of Bisecting k-means? I'm trying to cluster time
series of different length and I came up to an idea to use DTW as a
similarity measure, which seems to be adequate, but the thing is, I
cannot use it with K-means, since it's hard to define
Hello everyone.
I have a couple of questions.
1) Is there an implementation of DTW (Dynamic Time Warping) in Mahout
that could be used as a distance measure for clustering?
2) Why isn't there an implementation of K-mediods in Mahout? I'm
guessing that it could not be implemented efficiently
On Thu, Jan 8, 2015 at 7:00 AM, Marko Dinic marko.di...@nissatech.com
wrote:
1) Is there an implementation of DTW (Dynamic Time Warping) in Mahout that
could be used as a distance measure for clustering?
No.
2) Why isn't there an implementation of K-mediods in Mahout? I'm guessing
that
10 matches
Mail list logo