J/ApJ/820/8 3FGL sources statistical classifications (Saz Parkinson+, 2016)
Classification and ranking of Fermi LAT gamma-ray sources from the 3FGL catalog
using machine learning techniques.
Saz Parkinson P.M., Xu H., Yu P.L.H., Salvetti D., Marelli M., Falcone A.D.
<Astrophys. J., 820, 8 (2016)>
=2016ApJ...820....8S 2016ApJ...820....8S (SIMBAD/NED BibCode)
ADC_Keywords: Gamma rays ; Pulsars ; Active gal. nuclei
Keywords: gamma rays: stars; methods: statistical; pulsars: general
Abstract:
We apply a number of statistical and machine learning techniques to
classify and rank gamma-ray sources from the Third Fermi Large Area
Telescope Source Catalog (3FGL), according to their likelihood of
falling into the two major classes of gamma-ray emitters: pulsars
(PSR) or active galactic nuclei (AGNs). Using 1904 3FGL sources that
have been identified/associated with AGNs (1738) and PSR (166), we
train (using 70% of our sample) and test (using 30%) our algorithms
and find that the best overall accuracy (>96%) is obtained with the
Random Forest (RF) technique, while using a logistic regression (LR)
algorithm results in only marginally lower accuracy. We apply the same
techniques on a subsample of 142 known gamma-ray pulsars to classify
them into two major subcategories: young (YNG) and millisecond pulsars
(MSP). Once more, the RF algorithm has the best overall accuracy
(∼90%), while a boosted LR analysis comes a close second. We apply our
two best models (RF and LR) to the entire 3FGL catalog, providing
predictions on the likely nature of unassociated sources, including
the likely type of pulsar (YNG or MSP). We also use our predictions to
shed light on the possible nature of some gamma-ray sources with known
associations (e.g., binaries, supernova remnants/pulsar wind nebulae).
Finally, we provide a list of plausible X-ray counterparts for some
pulsar candidates, obtained using Swift, Chandra, and XMM. The results
of our study will be of interest both for in-depth follow-up searches
(e.g., pulsar) at various wavelengths and for broader population
studies.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table5.dat 156 3021 Results of our best models, as applied to
the 3FGL catalog
--------------------------------------------------------------------------------
See also:
J/ApJ/814/128 : Timing noise + astrometry of Fermi-LAT pulsars (Kerr+, 2015)
J/ApJ/810/14 : Third catalog of LAT-detected AGNs (3LAC) (Ackermann+, 2015)
J/ApJS/218/23 : Fermi LAT third source catalog (3FGL) (Acero+, 2015)
J/ApJS/208/17 : 2nd Fermi LAT cat. of gamma-ray pulsars (2PC) (Abdo+, 2013)
J/ApJ/756/33 : X-ray sources near 2 bright γ-ray sources (Cheung+, 2012)
J/MNRAS/424/2832 : Pulsars in γ-ray sources (Lee+, 2012)
J/MNRAS/424/L64 : AGN/pulsar distinction for 2FGL sources (Mirabal+, 2012)
J/ApJ/753/83 : Associations to 1FGL sources (Ackermann+, 2012)
J/ApJS/199/31 : Fermi LAT second source catalog (2FGL) (Nolan+, 2012)
J/ApJS/188/405 : Fermi-LAT first source catalog (1FGL) (Abdo+, 2010)
J/ApJS/183/46 : Fermi/LAT bright gamma-ray source list (0FGL) (Abdo+, 2009)
Byte-by-byte Description of file: table5.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 4 I4 --- Seq Running index number
6- 9 A4 --- --- [3FGL]
10- 22 A13 --- 3FGL Source 3FGL identifier (JHHMM.m+DDMMa)
24- 30 F7.2 --- Sig [4/1049] The 4-year significance
32- 39 E8.2 ct/cm2/MeV/s Flux Flux density; photons/cm2/MeV/s
41- 47 F7.3 deg RAdeg Right Ascension (J2000)
49- 56 F8.4 deg DEdeg Declination (J2000)
58- 63 F6.2 deg GLON Galactic longitude
65- 72 F8.4 deg GLAT Galactic latitude
74- 99 A26 --- ASSOC Likely associated source identifier
101-105 A5 --- Class 3FGL catalog class designation (1)
107-110 F4.2 --- LR-P Logistic regression (LR) probability
112-114 A3 --- LR-Pred LR predicted class
(2294 AGN & 727 PSR)
116-119 F4.2 --- RF-P Random Forest (RF) probability
121-123 A3 --- RF-Pred RF predicted class
(2368 AGN & 653 PSR)
125-130 F6.2 --- PSR-Out [-0.8/765] PSR outlyingness
132-138 F7.2 --- AGN-Out [-0.9/4912] AGN outlyingness
140-143 F4.2 --- BLR-PSR-P Boosted LR probability of being
a YNG pulsar
145-147 A3 --- BLR-PSR-Pred Boosted LR predicted class (363 MSP;
430 YNG)
149-152 F4.2 --- RF-PSR-P ? RF probability of being a YNG pulsar
154-156 A3 --- RF-PSR-Pred RF predicted pulsar class (361 MSP;
410 YNG)
--------------------------------------------------------------------------------
Note (1): Class as follows:
BCU = Blazar candidate of uncertain type (5 sources, 565 "bcu")
BIN = Binary (1 source)
BLL = BL Lac type of blazar (18 sources, 641 "bll")
FSRQ = FSRQ type of blazar (38 sources, 445 "fsrq")
GAL = Normal galaxy (or part) (2 sources, 1 "gal")
HMB = High-mass binary (3 sources)
NLSY1 = Narrow-line Seyfert 1 (2 sources, 3 "nlsy1")
NOV = Nova (1 source)
PSR = Pulsar, identified by pulsations (142 sources)
psr = Pulsar, no pulsations seen in LAT yet (24 sources)
PWN = Pulsar wind nebula (7 sources, 2 "pwn")
RDG = Radio galaxy (3 sources, 12 "rdg")
SNR = Supernova remnant (11 sources, 11 "snr")
agn = Non-blazar active galaxy (3 sources)
css = Compact steep spectrum quasar (1 source)
glc = Globular cluster (15 sources)
sbg = Starburst galaxy (4 sources)
sey = Seyfert galaxy (1 source)
spp = Supernova remnant/pulsar wind nebula (49 sources)
ssrq = Soft-spectrum radio quasar (3 sources)
See also the 3FGL: Acero+, 2015, J/ApJS/218/23
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Prepared by [AAS], Emmanuelle Perret [CDS] 30-May-2016