Share Email Print

Proceedings Paper

Classification of AGNs from stars and normal galaxies by surport vector machines
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

In order to explore the spectral energy distribution of various objects in a multidimensional parameter space, the multiwavelenghth data of quasars, BL Lacs, active galaxies, stars and normal galaxies are obtained by positional cross-identification, which are from optical(USNO A-2), X-ray(ROSAT), infrared(2MASS) bands. Different classes of X-ray emitters populate distinct regions of a multidimensional parameter space. In this paper, an automatic classification technique called Support Vector Machines(SVMs) is put forward to classify them using 7 parameters and 10 parameters. Finally the results show SVMs is an effective method to separate AGNs from stars and normal galaxies with data from optical, X-ray bands and with data from optical, X-ray, infrared bands. Furthermore, we conclude that to classify objects is influenced not only by the method, but also by the chosen wavelengths. Moreover it is evident that the more wavelengths we choose, the higher the accuracy is.

Paper Details

Date Published: 19 December 2002
PDF: 8 pages
Proc. SPIE 4847, Astronomical Data Analysis II, (19 December 2002); doi: 10.1117/12.460412
Show Author Affiliations
Yanxia Zhang, National Astronomical Observatories (China)
Chenzhou Cui, National Astronomical Observatories (China)
Yongheng Zhao, National Astronomical Observatories (China)

Published in SPIE Proceedings Vol. 4847:
Astronomical Data Analysis II
Jean-Luc Starck; Fionn D. Murtagh, Editor(s)

© SPIE. Terms of Use
Back to Top