Bayesian Classification Methods for Synoptic Sky Surveys

Details Bayesian Classification Methods for Synoptic Sky Surveys Bayesian Classification Methods for Synoptic Sky Surveys

May 2009

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2009aas...21440711d&link_type=abstract

American Astronomical Society, AAS Meeting #214, #407.11; Bulletin of the American Astronomical Society, Vol. 41, p.670

Statistics

Methodology

Scientific paper

Modern synoptic sky surveys present some novel challenges for automated classification of detected events. Reliable and robust real-time classification is essential in order to exploit the full scientific potential of these surveys, and conduct optimised follow-up observations. One example is an automated detection and removal of artifacts which could masquerade as transient events on the sky. But a much more interesting problem is automated, real-time classification of variable and transient events in terms of their astrophysical nature. This generally has to be done using sparse and heterogeneous measurements for individual events, both from the survey pipelines and existing archives, a situation very different from the more traditional classification problems such as the star-galaxy separation. Bayesian methodology is desirable and attractive for this task, since these methods can deal with missing parameters. We are building Bayesian network classifiers using colors, light-curves and spectra for variable and transient events. The data used are both from archival observations, as well as from our own follow-up of recent transients from the PQ and CRTS surveys. We will describe the details about the network architecture as well as the data used and possible extensions.

Affiliated with

Also associated with

No associations

Rating

Bayesian Classification Methods for Synoptic Sky Surveys does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Bayesian Classification Methods for Synoptic Sky Surveys, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Bayesian Classification Methods for Synoptic Sky Surveys will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1106681