Evaluating the classification of Fermi BCUs from the 4FGL Catalog Using Machine Learning

Abstract

The recently published fourth Fermi Large Area Telescope source catalog (4FGL) reports 5065 gamma-ray sources in terms of direct observational gamma-ray properties. Among the sources, the largest population is the Active Galactic Nuclei (AGN), which consists of 3137 blazars, 42 radio galaxies, and 28 other AGNs. The blazar sample comprises 694 flat-spectrum radio quasars (FSRQs), 1131 BL Lac-type objects (BL Lacs), and 1312 blazar candidates of an unknown type (BCUs). The classification of blazars is difficult using optical spectroscopy given the limited knowledge with respect to their intrinsic properties, and the limited availability of astronomical observations. To overcome these challenges, machine learning algorithms are being investigated as alternative approaches. Using the 4FGL catalog, a sample of 3137 Fermi blazars with 23 parameters is systematically selected. Three established supervised machine learning algorithms (random forests (RFs), support vector machines (SVMs), artificial neural networks (ANNs)) are employed to general predictive models to classify the BCUs. We analyze the results for all of the different combinations of parameters. Interestingly, a previously reported trend the use of more parameters leading to higher accuracy is not found. Considering the least number of parameters used, combinations of eight, 12 or 10 parameters in the SVM, ANN, or RF generated models achieve the highest accuracy (Accuracy $\simeq$ 91.8\%, or $\simeq$ 92.9\%). Using the combined classification results from the optimal combinations of parameters, 724 BL Lac type candidates and 332 FSRQ type candidates are predicted; however, 256 remain without a clear prediction.

Accepted for publication in ApJ. 14 pages, 2 figures, 6 tables

References in corpus (56)

• Unified Schemes for Radio-Loud Active Galactic Nuclei
• Hot Accretion Flows Around Black Holes
• Fermi Large Area Telescope First Source Catalog
• The Spectral Energy Distribution of Fermi bright blazars
• The Second Catalog of Active Galactic Nuclei Detected by the Fermi Large Area Telescope
• The Third Catalog of Active Galactic Nuclei Detected by the Fermi Large Area Telescope
• The First Catalog of Active Galactic Nuclei Detected by the Fermi Large Area Telescope
• Bright AGN Source List from the First Three Months of the Fermi Large Area Telescope All-Sky Survey
• Data Mining and Machine Learning in Astronomy
• Multiwavelength Observations of Strong Flares From the TeV-Blazar 1ES 1959+650
• The Fourth Catalog of Active Galactic Nuclei Detected by the Fermi Large Area Telescope
• The transition between BL Lac objects and Flat Spectrum Radio Quasars
• The Fermi blazars divide
• The Spectral Energy Distributions of Fermi Blazars
• X-Ray Spectral Variability Signatures of Flares in BL Lac Objects
• Radiation Mechanisms and Physical Properties of GeV-TeV BL Lac Objects
• Machine Learning in Astronomy: a practical overview
• A Statistical Approach to Recognizing Source Classes for Unassociated Sources in the First Fermi-LAT Catalog
• Classification and Ranking of Fermi LAT Gamma-ray Sources from the 3FGL Catalog using Machine Learning Techniques
• On the jet properties of the gamma-ray loud active galactic nuclei
• The jet-disc connection in AGN
• Intrinsic $\gamma$-ray luminosity, black hole mass, jet and accretion in Fermi blazars
• Constraints on minimum electron Lorentz factor and matter content of jets for a sample of bright Fermi blazars
• Fermi's Sibyl: Mining the gamma-ray sky for dark matter subhaloes
• Jet formation in BL Lacertae objects with different accretion modes
• The physical properties of Fermi BL Lac objects jets
• Blazar Flaring Patterns (B-FlaP): Classifying Blazar Candidates of Uncertain type in the third Fermi-LAT catalog by Artificial Neural Networks
• Flux and Photon Spectral Index Distributions of Fermi-LAT Blazars And Contribution To The Extragalactic Gamma-ray Background
• On the BL Lacertae objects/radio quasars and the FRI/II dichotomy
• Gamma-Ray Active Galactic Nucleus Type through Machine-Learning Algorithms
• Search for gamma-ray-emitting active galactic nuclei in the Fermi-LAT unassociated sample using machine learning
• The Accretion Rates and Spectral Energy Distributions of BL Lacertae Objects
• Implications on the blazar sequence and inverse Compton models from Fermi bright blazars
• ASCA and contemporaneous ground-based observations of the BL Lacertae objects 1749+096 and 2200+420 (BL Lac)
• Research and characterisation of blazar candidates among the Fermi/LAT 3FGL catalogue using multivariate classifications
• 3FGLzoo. Classifying 3FGL Unassociated Fermi-LAT Gamma-ray Sources by Artificial Neural Networks
• Discerning the $\gamma$-ray emitting region in the flat spectrum radio quasars
• Clustering of gamma-ray burst types in the Fermi-GBM catalogue: indications of photosphere and synchrotron emissions during the prompt phase
• Evaluating the optical classification of Fermi BCUs using machine learning
• Evidence for the Secondary Emission as the Origin of Hard Spectra in TeV Blazars
• Basic properties of Fermi blazars and the "blazar sequence"
• Bethe-Heitler cascades as a plausible origin of hard spectra in distant TeV blazars
• On the origin of the soft photons of the high synchrotron peaked blazar : PKS 1424+240
• Multi-color optical monitoring of Mrk 501 from 2010 to 2015
• How to Find Variable Active Galactic Nuclei with Machine Learning
• What Powers the Most Relativistic Jets? II: Flat Spectrum Radio Quasars
• Foreword to the Focus Issue on Machine Learning in Astronomy and Astrophysics
• Accretion Disk MHD Winds and Blazar Classification
• A Determination of the Gamma-ray Flux and Photon Spectral Index Distributions of Blazars from the Fermi-LAT 3LAC
• Comparison between TeV and non-TeV BL Lac Objects
• The beaming effect and $\gamma$-ray emission for Fermi blazars
• Constraints on the location of the gamma-ray emission region for the gamma-ray-loud radio source GB 1310+487
• FSRQ/BL Lac dichotomy as the magnetized advective accretion process around black holes: a unified classification of blazars
• What determines the observational differences of blazars?
• The Intrinsic $\gamma$-ray Emissions of $Fermi$ Blazars
• On the intrinsic shape of gamma-ray spectrum for Fermi blazars

Cited by in corpus (10)

• Searching for AGN and Pulsar Candidates in 4FGL Unassociated Sources Using Machine Learning
• Searching for TeV Candidates in 4LAC High-synchrotron-peaked Frequency BL Lac Objects
• Using Neural Networks to Differentiate Newly Discovered BL Lacs and FSRQs among the 4FGL Unassociated Sources Employing Gamma-ray, X-ray, UV/Optical and IR Data
• Classification of blazar candidates of unknown type in Fermi 4LAC by unanimous voting from multiple Machine Learning Algorithms
• Classification of the \emph{Fermi}-LAT Blazar Candidates of Uncertain type using eXtreme Gradient Boosting
• Deep Learning Blazar Classification based on Multi-frequency Spectral Energy Distribution Data
• Exploring TeV candidates of Fermi blazars through machine learning
• Fermi LAT AGN classification using supervised machine learning
• An empirical "high-confidence" candidate zone for $Fermi$ BL Lacertae objects
• Estimation of redshift and associated uncertainty of Fermi/LAT extra-galactic sources with Deep Learning