Automatic classification of time-variable X-ray sources

Abstract

To maximize the discovery potential of future synoptic surveys, especially in the field of transient science, it will be necessary to use automatic classification to identify some of the astronomical sources. The data mining technique of supervised classification is suitable for this problem. Here, we present a supervised learning method to automatically classify variable X-ray sources in the second \textit{XMM-Newton} serendipitous source catalog (2XMMi-DR2). Random Forest is our classifier of choice since it is one of the most accurate learning algorithms available. Our training set consists of 873 variable sources and their features are derived from time series, spectra, and other multi-wavelength contextual information. The 10-fold cross validation accuracy of the training data is $\sim$97% on a seven-class data set. We applied the trained classification model to 411 unknown variable 2XMM sources to produce a probabilistically classified catalog. Using the classification margin and the Random Forest derived outlier measure, we identified 12 anomalous sources, of which, 2XMM J180658.7$-$500250 appears to be the most unusual source in the sample. Its X-ray spectra is suggestive of a ULX but its variability makes it highly unusual. Machine-learned classification and anomaly detection will facilitate scientific discoveries in the era of all-sky surveys.

Accepted for ApJ

References in corpus (38)

• The Wide-field Infrared Survey Explorer (WISE): Mission Description and Initial On-orbit Performance
• The generalised Lomb-Scargle periodogram. A new formalism for the floating-mean and Keplerian periodograms
• SUMSS: A Wide-Field Radio Imaging Survey of the Southern Sky. II. The Source Catalogue
• An Intermediate-mass Black Hole of Over 500 Solar Masses in the Galaxy ESO 243-49
• The XMM-Newton Serendipitous Survey. V. The Second XMM-Newton Serendipitous Source Catalogue
• Bayesian Blocks, A New Method to Analyze Structure in Photon Counting Data
• On Machine-Learned Classification of Variable Stars with Sparse and Noisy Time-Series Data
• The Molonglo Galactic Plane Survey (MGPS-2): Compact Source Catalogue
• Large Synoptic Survey Telescope: Overview
• Kepler Eclipsing Binary Stars. III. Classification of Kepler Eclipsing Binary Light Curves with Locally Linear Embedding
• Random forest automated supervised classification of Hipparcos periodic variable stars
• Automating Discovery and Classification of Transients and Variable Stars in the Synoptic Survey Era
• Probabilistic Cross-Identification of Astronomical Sources
• Exploring the Variable Sky with LINEAR. III. Classification of Periodic Light Curves
• VAST: An ASKAP Survey for Variables and Slow Transients
• The first outburst of the new magnetar candidate SGR 0501+4516
• Classification of X-ray Sources in the XMM-Newton Serendipitous Source Catalog
• RXJ0806.3+1527: a double degenerate binary with the shortest known orbital period (321s)
• The Dynamic Radio Sky
• A deep XMM-Newton observation of the ultraluminous X-ray source HoII X-1 : the case against a 1000 solar mass black hole
• Testing the Paradigm that Ultraluminous X-ray Sources as a Class Represent Accreting Intermediate-Mass Black Holes
• Cross-correlation of the 2XMMi catalogue with Data Release 7 of the Sloan Digital Sky Survey
• A ROSAT PSPC X-Ray Survey of the Small Magellanic Cloud
• Hard X-ray flares in IGR J08408-4503 unveil clumpy stellar winds
• Automated classification of Hipparcos unsolved variables
• The Photometric Classification Server for Pan-STARRS1
• Optical identification of XMM sources in the CFHTLS
• A New Sample of Candidate Intermediate-Mass Black Holes Selected by X-ray Variability
• X-ray source variability study of the M 31 central field using Chandra HRC-I
• Photometric Selection of QSO Candidates From GALEX Sources
• Optical identification of X-ray source 1RXS J180431.1-273932 as a magnetic cataclysmic variable
• Two Rapidly Variable Galactic X-ray Transients Observed with Chandra, XMM and Suzaku
• AX J1910.7+0917 and three newly discovered INTEGRAL sources
• A New 626 s Periodic X-ray Source in the Direction of the Galactic Center
• Long-term X-ray Variability Study of IC342 from XMM-Newton Observations
• Flashes in a Star Stream: Automated Classification of Astronomical Transient Events
• XMMSL1 J060636.2-694933: An XMM-Newton Slew discovery and Swift/Magellan follow up of a new Classical Nova in the LMC
• XMM-Newton confirmation of a new intermediate polar: XMMU J185330.7-012815

Cited by in corpus (22)

• X-ray Emissions from Accreting White Dwarfs: a Review
• Machine-learning Approaches to Exoplanet Transit Detection and Candidate Validation in Wide-field Ground-based Surveys
• Probabilistic classification of X-ray sources applied to Swift-XRT and XMM-Newton catalogs
• Classifying Unidentified X-ray Sources in the Chandra Source Catalog Using a Multiwavelength Machine-learning Approach
• Quantifying ionospheric effects on time-domain astrophysics with the Murchison Widefield Array
• Supervised Ensemble Classification of Kepler Variable Stars
• The EXOD search for faint transients in XMM-Newton observations: Method and discovery of four extragalactic Type I X-ray Bursters
• Unsupervised Machine Learning for the Classification of Astrophysical X-ray Sources
• Identifying New X-ray Binary Candidates in M31 using Random Forest Classification
• X-ray Sources in the Dwarf Spheroidal Galaxy Draco
• Selection of Burst-like Transients and Stochastic Variables Using Multi-Band Image Differencing in the Pan-STARRS1 Medium-Deep Survey
• On the optical counterparts of radio transients and variables
• 3D mapping of the neutral X-ray absorption in the local interstellar medium: The Gaia and XMM-Newton synergy
• Searching for Quasi-Periodic Eruptions using Machine Learning
• On the use of variability time-scales as an early classifier of radio transients and variables
• X-ray Sources Classification Using Machine Learning: A Study with EP-WXT Pathfinder LEIA
• The time-variable ultraviolet sky: Active Galactic Nuclei, Stars and White Dwarfs
• Automated algorithms to build Active Galactic Nuclei classifiers
• State Space Modelling for detecting and characterising Gravitational Waves afterglows
• SRG/ART-XC all-sky X-ray survey: Catalog of sources detected during the first year
• XClass: An Automated Multiwavelength Machine-Learning Pipeline for Classification of Extragalactic X-ray Sources. I. Pipeline Description
• Search for Astrophysical Transients on Limiting Time Scales and Their Classification Based on INTEGRAL Data