[R] Classification of supernovae - a challenge

[Jarle Brinchmann]

Hi all,

I thought I'd just point out, to those not having yet seen this, that
today there was a classification challenge posted for astronomy.

The web-site is http://www.hep.anl.gov/SNchallenge/

[I have nothing to do with this project so don't ask me any details!]

Basically the idea behind is that future surveys of the sky will turn
up a large number of supernovae (SNe). The surveys' main goal is to
try to understand the mysterious dark energy which seems to make up
~70% of the energy density of the Universe. The number of these SNe
that will be detected is expected to be moderately large (~10^5). In
the past astronomers have studied these using spectroscopic data which
allow you to accurately classify supernovae but that will not be
possible in the future. Instead one will have to rely on measurements
of flux in broad bands to classify supernovae.

This challenge then is to try to classify SNe using photometry only
and they have provided training & test data on the web sites above.
The characteristic features are typically encoded in the shape of the
flux curve as a function of time but other aspects might be useful
too. There is a description on the web site for more information.

Anyway, just thought some of you might find this a fun challenge for
you or your students - the deadline is May 1.

      Cheers,
           Jarle.