October 5-9, 2014


P1.5 Knowledge Discovery in Mega-Spectra Archives

Petr Škoda (Astronomical Institute of the Academy of Sciences)

L. Lopatovský, A. Palička (Faculty of Information Technology, Czech Technical University in Prague), P. Bromová (Faculty of Information Technology, Brno University of Technology), J. Vážný (Astronomical Institute of the Academy of Sciences)

The recent progress of astronomical instrumentation resulted in the construction of multi-object spectrographs with hundreds to thousands of micro-slits or optical fibers allowing the acquisition of tens of thousands of spectra of celestial objects per observing night. Currently there are several spectroscopic surveys (as SDSS or LAMOST) containing millions of spectra and much larger are in preparation.

These surveys are being processed by automatic pipelines, spectrum by spectrum, in order to estimate physical parameters of individual objects resulting in extensive catalogues, used typically to construct the better models of space-kinematic structure and evolution of the Universe or its subsystems. Such surveys are, however, very good source of homogenized, pre-processed data for application of machine learning techniques common in Astroinformatics.

We present challenges of knowledge discovery in such surveys as well as practical examples of machine learning based on specific shapes of spectral features used in searching for new candidates of interesting astronomical objects, namely Be and B[e] stars and quasars.

Finally, the need for the new Big Data approach in such effort will be stressed, including the development of new massively parallel machine learning algorithms as well as better collaboration with non-astronomical communities sharing similar problems (e.g. in Earth observation).

Mode of presentation: poster

Applicable ADASS XXIV theme category: Big Data Challenges