A Comparative Study of Machine-learning Methods for X-Ray Binary Classification

Beurs, Zoë L. de; Islam, Nazma; Gopalan, Giri; Vrtilek, S. D.

doi:10.3847/1538-4357/ac6184

Cited by 7 publications

(3 citation statements)

References 40 publications

(46 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The SIMBAD catalog is overall accurate and has been used carefully to verify source classifications previously (e.g., de Beurs et al 2022;Li et al 2022). While constructing our TD, we find that some sources have wrong SIMBAD classifications.…”

Section: Comparison To Catalogs With Known Source Classesmentioning

confidence: 85%

Classifying Unidentified X-Ray Sources in the Chandra Source Catalog Using a Multiwavelength Machine-learning Approach

Yang

Hare

Kargaltsev

et al. 2022

ApJ

View full text Add to dashboard Cite

The rapid increase in serendipitous X-ray source detections requires the development of novel approaches to efficiently explore the nature of X-ray sources. If even a fraction of these sources could be reliably classified, it would enable population studies for various astrophysical source types on a much larger scale than currently possible. Classification of large numbers of sources from multiple classes characterized by multiple properties (features) must be done automatically and supervised machine learning (ML) seems to provide the only feasible approach. We perform classification of Chandra Source Catalog version 2.0 (CSCv2) sources to explore the potential of the ML approach and identify various biases, limitations, and bottlenecks that present themselves in these kinds of studies. We establish the framework and present a flexible and expandable Python pipeline, which can be used and improved by others. We also release the training data set of 2941 X-ray sources with confidently established classes. In addition to providing probabilistic classifications of 66,369 CSCv2 sources (21% of the entire CSCv2 catalog), we perform several narrower-focused case studies (high-mass X-ray binary candidates and X-ray sources within the extent of the H.E.S.S. TeV sources) to demonstrate some possible applications of our ML approach. We also discuss future possible modifications of the presented pipeline, which are expected to lead to substantial improvements in classification confidences.

show abstract

Section: Comparison To Catalogs With Known Source Classesmentioning

confidence: 85%

Classifying Unidentified X-Ray Sources in the Chandra Source Catalog Using a Multiwavelength Machine-learning Approach

Yang

Hare

Kargaltsev

et al. 2022

ApJ

View full text Add to dashboard Cite

show abstract

“…Analysis of short scans on each XRB taken on the first 3 days was done using the complex fringe visibility function as measured at each u,v point (baseline and time). a Indicates sources with sufficient frequency ranges to analyze the SED in Section 3. b While pulsations have not been detected, the position of 4U 1700-37 in color-color-intensity diagrams locates it as a pulsar (Vrtilek & Boroson 2013;Gopalan et al 2015;de Beurs et al 2022).…”

Section: Radio Observationsmentioning

confidence: 99%

Spectral Energy Distributions of Southern Binary X-Ray Sources

Dickey,

Vrtilek,

McCollough

et al. 2023

ApJS

Self Cite

View full text Add to dashboard Cite

The rapid variability of X-ray binaries (XRBs) produces a wide range of X-ray states that are linked to activity across the electromagnetic spectrum. It is particularly challenging to study a sample of sources large enough to include all types in their various states, and to cover the full range of frequencies that show flux density variations. Simultaneous observations with many telescopes are necessary. In this project, we monitor 48 XRBs with seven telescopes across the electromagnetic spectrum from 5 × 109 to 1019 Hz, including ground-based radio, IR, and optical observatories, and five instruments on two spacecraft over a 1 week period. We construct spectral energy distributions and matching X-ray color–intensity diagrams for 20 sources that have the most extensive detections. Our observations are consistent with several models of expected behavior proposed for the different classes: we detect no significant radio emission from pulsars or atoll sources, but we do detect radio emission from Z sources in the normal or horizontal branch, and from black holes in the high/soft, low/hard, and quiescent states. The survey data provide useful constraints for more detailed models predicting behavior from the different classes of sources.

show abstract

“…However, the abovementioned studies predominantly use machine-learning methods of the unsupervised type, where only the observed features of the GRBs are inputted into the models, but not the labels (the GRBs' physical classes being Type I or II). On the other hand, the other type of machinelearning methods, supervised methods, are also commonly employed by astronomy researchers in the classification of other astronomical objects (e.g., Luo et al 2023;Zhu-Ge et al 2023;Connor & van Leeuwen 2018;Butter et al 2022;Coronado-Blázquez 2022;de Beurs et al 2022;Fan et al 2022;Villa-Ortega et al 2022;Yang et al 2022a;Kaur et al 2023), although studies on the application of supervised methods on GRB are scarce. Since supervised methods take both features and labels as input, and can produce deterministic predictions of the class of new GRBs, they can be helpful in identifying the true physical origin of intermingled GRBs.…”

Section: Introductionmentioning

confidence: 99%

Identifying the Physical Origin of Gamma-Ray Bursts with Supervised Machine Learning

Luo,

Wang,

Zhu-Ge

et al. 2023

ApJ

View full text Add to dashboard Cite

The empirical classification of gamma-ray bursts (GRBs) into long and short GRBs based on their durations is already firmly established. This empirical classification is generally linked to the physical classification of GRBs originating from compact binary mergers and GRBs originating from massive star collapses, or Type I and II GRBs, with the majority of short GRBs belonging to Type I and the majority of long GRBs belonging to Type II. However, there is a significant overlap in the duration distributions of long and short GRBs. Furthermore, some intermingled GRBs, i.e., short-duration Type II and long-duration Type I GRBs, have been reported. A multiparameter classification scheme of GRBs is evidently needed. In this paper, we seek to build such a classification scheme with supervised machine-learning methods, chiefly XGBoost. We utilize the GRB Big Table and Greiner’s GRB catalog and divide the input features into three subgroups: prompt emission, afterglow, and host galaxy. We find that the prompt emission subgroup performs the best in distinguishing between Type I and II GRBs. We also find the most important distinguishing features in prompt emission to be T 90, the hardness ratio, and fluence. After building the machine-learning model, we apply it to the currently unclassified GRBs to predict their probabilities of being either GRB class, and we assign the most probable class of each GRB to be its possible physical class.

show abstract

A Comparative Study of Machine-learning Methods for X-Ray Binary Classification

Cited by 7 publications

References 40 publications

Classifying Unidentified X-Ray Sources in the Chandra Source Catalog Using a Multiwavelength Machine-learning Approach

Classifying Unidentified X-Ray Sources in the Chandra Source Catalog Using a Multiwavelength Machine-learning Approach

Spectral Energy Distributions of Southern Binary X-Ray Sources

Identifying the Physical Origin of Gamma-Ray Bursts with Supervised Machine Learning

Contact Info

Product

Resources

About