J/ApJ/813/28 Autoclassification of the variable 3XMM sources (Farrell+, 2015)
Autoclassification of the variable 3XMM sources using the random forest machine
learning algorithm.
Farrell S.A., Murphy T., Lo K.K.
<Astrophys. J., 813, 28 (2015)>
=2015ApJ...813...28F 2015ApJ...813...28F (SIMBAD/NED BibCode)
ADC_Keywords: X-ray sources
Keywords: catalogs; methods: statistical
Abstract:
In the current era of large surveys and massive data sets,
autoclassification of astrophysical sources using intelligent
algorithms is becoming increasingly important. In this paper we
present the catalog of variable sources in the Third XMM-Newton
Serendipitous Source catalog (3XMM) autoclassified using the Random
Forest machine learning algorithm (RF). We used a sample of manually
classified variable sources from the second data release of the
XMM-Newton catalogs (2XMMi-DR2) to train the classifier, obtaining an
accuracy of ∼92%. We also evaluated the effectiveness of identifying
spurious detections using a sample of spurious sources, achieving an
accuracy of ∼95%. Manual investigation of a random sample of
classified sources confirmed these accuracy levels and showed that the
Random Forest machine learning algorithm is highly effective at
automatically classifying 3XMM sources. Here we present the catalog of
classified 3XMM variable sources. We also present three previously
unidentified unusual sources that were flagged as outlier sources by
the algorithm: a new candidate supergiant fast X-ray transient, a 400s
X-ray pulsar, and an eclipsing 5hr binary system coincident with a
known Cepheid.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table4.dat 86 2876 3XMM variable source classifications using the Random
Forest machine learning algorithm (RF)
--------------------------------------------------------------------------------
See also:
B/xmm : XMM-Newton Observation Log (XMM-Newton Science Operation Center, 2012)
IX/46 : XMM-Newton Serendipitous Source Catalogue 3XMM-DR5 (XMM-SSC, 2016)
IX/44 : XMM-Newton Serendipitous Source Catalogue 3XMM-DR4 (XMM-SSC, 2013)
J/MNRAS/414/2602 : Automated classification of HIP variables (Dubath+, 2011)
J/A+A/493/339 : XMM-Newton serendipitous Survey. V. (Watson+, 2009)
Byte-by-byte Description of file: table4.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 4 A4 --- --- [3XMM]
6- 21 A16 --- 3XMM 3XMM IAU designation (JHHMMSS.s+DDMMSS)
23- 27 F5.3 --- P(AGN) RF classifier probability; AGN source class
29- 33 F5.3 --- P(CV) RF classifier probability; CV source class
35- 39 F5.3 --- P(GRB) RF classifier probability; GRB source class
41- 45 F5.3 --- P(STAR) RF classifier probability; STAR source class
47- 51 F5.3 --- P(ULX) RF classifier probability; ULX source class
53- 57 F5.3 --- P(XRB) RF classifier probability; XRB source class
59- 63 F5.3 --- P(Max) Max RF classifier probability over all
source classes
65- 69 F5.3 --- P(Spur) RF classifier probability that the source
is spurious
71- 74 A4 --- Cl Source class assigned by RF classifier (1)
76- 79 I4 --- Out [-1/2891] Outlier measure of the source (2)
81- 86 F6.3 --- Margin [-0.6/1] Classification margin of the source (3)
--------------------------------------------------------------------------------
Note (1): Source class:
STAR = 1942 sources
XRB = 579 sources (X-ray binary)
AGN = 144 sources (active galactic nucleus)
CV = 152 sources (cataclysmic variable)
ULX = 54 sources (ultraluminous X-ray source)
GRB = 5 sources (gamma-ray burst)
Note (2): Outlier measure of the source. Equation 10 in Lo et al.
(2014ApJ...786...20L 2014ApJ...786...20L) provides a definition of this parameter.
Larger values indicate a higher likelihood of being an outlier.
Note (3): Margin=2*PMax-1
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Prepared by [AAS], Emmanuelle Perret [CDS] 11-Feb-2016