Abstract
The Discrete Source Classifier in the Gaia Astrophysical parameters processing chain has the task of sorting the 109 detected Gaia sources into broad astrophysical classes. These classifications will form part of the final database and are used to trigger various parameterizing algorithms further down the chain. Available input information consists of low-resolution spectra from the Gaia photometers, sky position and apparent magnitude, proper motion and parallax measurements, possible variability information and, for the bright sources, high-resolution spectra from the region 8470–8740 Å, although not all of this information is currently used. Since the classification scheme must deal with several different types of input data, we use a modular approach to classify probabilistically each type of input data and then combine these probabilities. Specific problems facing the DSC include the imbalance in the class fractions, with the vast majority of the sources being single or multiple stars and a small minority falling into classes such as quasar-stellar objects or galaxies. The classifier must deal with a wide range of noise characteristics and also be robust against missing or damaged data. I will describe the system in overview and then go through some of these problems, and our responses to them, in more detail.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer Science+Business Media New York
About this chapter
Cite this chapter
Smith, K.W. (2012). The Discrete Source Classifier in Gaia-Apsis. In: Sarro, L., Eyer, L., O'Mullane, W., De Ridder, J. (eds) Astrostatistics and Data Mining. Springer Series in Astrostatistics, vol 2. Springer, New York, NY. https://doi.org/10.1007/978-1-4614-3323-1_25
Download citation
DOI: https://doi.org/10.1007/978-1-4614-3323-1_25
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4614-3322-4
Online ISBN: 978-1-4614-3323-1
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)