CLASSIFICATION and RANKING of FERMI LAT GAMMA-RAY SOURCES from the 3FGL CATALOG USING MACHINE LEARNING TECHNIQUES

P. M. Saz Parkinson, H. Xu, P. L.H. Yu, D. Salvetti, M. Marelli, Abraham David Falcone

Research output: Contribution to journalArticle

29 Citations (Scopus)

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.

Original languageEnglish (US)
Article number8
JournalAstrophysical Journal
Volume820
Issue number1
DOIs
StatePublished - Mar 20 2016

Fingerprint

machine learning
pulsars
catalogs
logistics
prediction
train
regression analysis
wavelength
active galactic nuclei
gamma rays
supernova remnants
XMM-Newton telescope
nebulae
predictions
falling
lists
learning
young
emitters
telescopes

All Science Journal Classification (ASJC) codes

  • Astronomy and Astrophysics
  • Space and Planetary Science

Cite this

Saz Parkinson, P. M. ; Xu, H. ; Yu, P. L.H. ; Salvetti, D. ; Marelli, M. ; Falcone, Abraham David. / CLASSIFICATION and RANKING of FERMI LAT GAMMA-RAY SOURCES from the 3FGL CATALOG USING MACHINE LEARNING TECHNIQUES. In: Astrophysical Journal. 2016 ; Vol. 820, No. 1.
@article{b32ca3d059444252898acfd85350fba8,
title = "CLASSIFICATION and RANKING of FERMI LAT GAMMA-RAY SOURCES from the 3FGL CATALOG USING MACHINE LEARNING TECHNIQUES",
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.",
author = "{Saz Parkinson}, {P. M.} and H. Xu and Yu, {P. L.H.} and D. Salvetti and M. Marelli and Falcone, {Abraham David}",
year = "2016",
month = "3",
day = "20",
doi = "10.3847/0004-637X/820/1/8",
language = "English (US)",
volume = "820",
journal = "Astrophysical Journal",
issn = "0004-637X",
publisher = "IOP Publishing Ltd.",
number = "1",

}

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, Abraham David.

In: Astrophysical Journal, Vol. 820, No. 1, 8, 20.03.2016.

Research output: Contribution to journalArticle

TY - JOUR

T1 - CLASSIFICATION and RANKING of FERMI LAT GAMMA-RAY SOURCES from the 3FGL CATALOG USING MACHINE LEARNING TECHNIQUES

AU - Saz Parkinson, P. M.

AU - Xu, H.

AU - Yu, P. L.H.

AU - Salvetti, D.

AU - Marelli, M.

AU - Falcone, Abraham David

PY - 2016/3/20

Y1 - 2016/3/20

N2 - 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.

AB - 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.

UR - http://www.scopus.com/inward/record.url?scp=84962243488&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962243488&partnerID=8YFLogxK

U2 - 10.3847/0004-637X/820/1/8

DO - 10.3847/0004-637X/820/1/8

M3 - Article

AN - SCOPUS:84962243488

VL - 820

JO - Astrophysical Journal

JF - Astrophysical Journal

SN - 0004-637X

IS - 1

M1 - 8

ER -