Searching for AGN and pulsar candidates in 4FGL unassociated sources using machine learning

Abstract

In the fourth Fermi Large Area Telescope source catalog (4FGL), 5064 γ-ray sources are reported, including 3207 active galactic nuclei (AGNs), 239 pulsars, 1336 unassociated sources, 92 sources with weak association with blazars at low Galactic latitudes and 190 other sources. We employ two different supervised machine learning classifiers, combined with the direct observation parameters given by the 4FGL fits table, to search for sources potentially classified as AGNs and pulsars in the 1336 unassociated sources. In order to reduce the error caused by the large difference in the sizes of samples, we divide the classification process into two separate steps in order to identify the AGNs and the pulsars. First, we select the identified AGNs from all of the samples, and then select the identified pulsars from the remaining cases. Using the 4FGL sources associated or identified as AGNs, pulsars and other sources with the features selected through the K-S test and the random forest (RF) feature importance measurement, we trained, optimized and tested our classifier models. Then, the models are applied to classify the 1336 unassociated sources. According to the calculation results of the two classifiers, we report the sensitivity, specificity, accuracy in each step and the class of unassociated sources given by each classifier. The accuracy obtained in the first step is approximately 95%; in the second step, the obtained overall accuracy is approximately 80%. Combining the results of the two classifiers, we predict that there are 583 AGN-type candidates, 115 pulsar-type candidates, 154 other types of γ-ray candidates and 484 of uncertain types.

Machine learning methods for constructing probabilisticFermi-LAT catalogs

“…Classification of Fermi -LAT 4FGL sources into three classes has been considered earlier by, for example, Zhu et al (2021), who primarily use a two-step classification procedure, where in the first step AGNs are separated from the rest of sources and in the second step the remaining sources are split into pulsars and OTHER sources. Zhu et al (2021) have also tested a simultaneous classification of sources into three classes (AGN, pulsars, and OTHER), but the results are inconsistent for the two ML algorithms used by Zhu et al (2021) (RF and NN). In particular, the number of OTHER sources predicted by NN is zero.…”

“…Bayesian association probabilities were also included in the 4FGL catalog (Abdollahi et al 2020) for faint sources. Probabilistic classification of unassociated Fermi -LAT sources was performed by, for example, Ackermann et al (2012), Saz Parkinson et al (2016, Mirabal et al (2016), Lefaucheur & Pita (2017), Luo et al (2020), Finke et al (2021), and Zhu et al (2021), or in the application for the subclassification of blazars by Hassan et al (2013), Doert & Errando (2014), Chiaro et al (2016), Salvetti et al (2017), and Kovačević et al (2019Kovačević et al ( , 2020, and in subclassification of pulsars by Lee et al (2012) and Saz Parkinson et al (2016). In this work we considered the classification of gamma-ray sources into two classes -active galactic nuclei (AGNs) and pulsars -as well as into three classes -AGNs, pulsars, and other associated sources ("OTHER").…”

Searching for TeV Candidates in 4LAC High-synchrotron-peaked Frequency BL Lac Objects

“…2. PTC SELECTION USING ML ML techniques are popular in the field of astronomical data mining and data analysis (Ball & Brunner 2010;Mirabal et al 2012;Chiaro et al 2016;Saz Parkinson et al 2016;Salvetti et al 2017;Baron 2019;Kang et al 2019a,b;Arsioli & Dedin 2020;Fraga et al 2021;Zhu et al 2021). According to whether the classification of a sample is given, ML can be mainly divided into supervised machine-learning (SML) and unsupervised machine-learning (USML) techniques (Baron 2019).…”

“…When evaluating a classifier, there are different metrics for performance evaluation that can be used, such as accuracy, recall, precision, and the receiver operating characteristic (ROC) curve (Baron 2019). Accuracy is widely used, but simple classification accuracy is usually dominated by large samples from imbalanced data sets (Zhu et al 2021). In this study, we used the balanced-accuracy (i.e., the average accuracy of positive and negative samples) rather than simple accuracy to evaluate the classifier performance.…”

Multiwavelength Spectral Analysis and Neural Network Classification of Counterparts to 4FGL Unassociated Sources

“…Searching in the radio band is similarly difficult, requiring assumptions to link gamma-ray and radio emission without interpreting the intermediate wavelengths, but the detection of characteristic radio pulsations is a direct route to locating pulsars and some previous works have used radio properties to predict pulsar membership after efficient searches (for example, Frail et al 2018). Recently, Zhu et al (2021) conducted ML on the 4FGL unassociated sources, but restricted their analysis to only gamma-ray properties. Analysis of X-ray observations specifically can capture the synchrotron peaks of non-thermal emitters like blazars, a valuable region for discriminating the spectra of pulsars and blazars.…”