[ml] jung, graph clustering

Mike Schachter mike at mindmech.com
Tue Jun 1 18:43:58 PDT 2010


Hey Andreas,

No problem delaying my presentation plans, but I'd appreciate it if you all
would keep the wiki updated and keep more of your collaborations on the
public list so everybody knows what's going on, especially me! There was
no indication that anyone had specific plans for tomorrow night's meeting,
which is why I planned on presenting something.

  mike




On Tue, Jun 1, 2010 at 6:34 PM, Andreas von Hessling
<vonhessling at gmail.com>wrote:

> Hi Mike,
> could you delay your plan for a week?  This is the last official ML
> session before the submission deadline; we want to use it to make sure
> we can generate a submission for the contest.
> Next week you can have as much time as you want;  I've personally also
> been interested in JUNG since I ran across it.
> Thanks!
>
> Andy
>
>
> On Tue, Jun 1, 2010 at 6:15 PM, Mike Schachter <mike at mindmech.com> wrote:
> > Oh and sorry about not setting up an FTP server yet, will get on that
> > first thing tomorrow morning.
> >
> >
> >
> > On Tue, Jun 1, 2010 at 6:11 PM, Mike Schachter <mike at mindmech.com>
> wrote:
> >>
> >> Hello everyone!
> >>
> >> I haven't gotten anywhere with clustering via self-organizing
> >> feature maps, but went off on a tangent after reading a section
> >> of Duda's book on machine learning (which is a classic that you
> >> should probably buy):
> >>
> >>
> >>
> http://www.amazon.com/Pattern-Classification-2nd-Richard-Duda/dp/0471056693
> >>
> >> In the chapter on clustering they talk briefly about clustering
> >> using graph algorithms, which I think Erin mentioned at a meeting
> >> a few weeks ago. I worked once with a pretty sweet graph modeling
> >> tool call JUNG and am going to try and implement the approach to
> >> create a visualizable cluster of the skill sets. Link to JUNG:
> >>
> >> http://jung.sourceforge.net/
> >>
> >> Tomorrow I'll go over the graphical approach real quick, along with
> >> a neural network in the Duda/Hart book that can be used for
> dimensionality
> >> reduction, which could be promising.
> >>
> >> See you tomorrow! Looking forward to it,
> >>
> >>    mike
> >>
> >
> >
> > _______________________________________________
> > ml mailing list
> > ml at lists.noisebridge.net
> > https://www.noisebridge.net/mailman/listinfo/ml
> >
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.noisebridge.net/pipermail/ml/attachments/20100601/827aaf95/attachment.htm 


More information about the ml mailing list