New results in applying the machine learning to GRB redshift estimation

Istvn I. Racz, Dezso Ribli, Z. Bagoly, Istvn Csabai, Istvn Horvath, Lajos G. Balzs

Research output: Contribution to journalConference article

Abstract

Gamma-ray bursts (GRBs) are the most energetic transients in the far Universe. Several thousands of GRBs have been observed so far but we could measure the distance of only a few hundreds. We studied the parameters of GRBs with available spectroscopic redshift in order to be able to estimate the redshift of those GRBs without a measured one. To calculate their distances we applied two machine-learning estimator methods: random forest regressor and XGBoost. For the process we used selected gamma, x-ray and ultraviolet parameters from the Swift GRB catalog, which contains the measured spectroscopic redshift of 328 GRBs. We found a significantly higher correlation between the measured and estimated redshift, we have improved the correlation in multiple steps from 0.57 (published by Ukwatta et al., 2016) to 0.67. It seems that both the random forest and the XGBoost methods give similarly high correlation. For further improvements additional redshift measurements are required.

Original languageEnglish
JournalProceedings of Science
Volume2017-October
Publication statusPublished - Jan 1 2017
Event7th International Fermi Symposium, IFS 2017 - Garmisch-Partenkirchen, Germany
Duration: Oct 15 2017Oct 20 2017

Fingerprint

machine learning
gamma ray bursts
estimators
catalogs
universe
estimates
x rays

ASJC Scopus subject areas

  • General

Cite this

Racz, I. I., Ribli, D., Bagoly, Z., Csabai, I., Horvath, I., & Balzs, L. G. (2017). New results in applying the machine learning to GRB redshift estimation. Proceedings of Science, 2017-October.

New results in applying the machine learning to GRB redshift estimation. / Racz, Istvn I.; Ribli, Dezso; Bagoly, Z.; Csabai, Istvn; Horvath, Istvn; Balzs, Lajos G.

In: Proceedings of Science, Vol. 2017-October, 01.01.2017.

Research output: Contribution to journalConference article

Racz, II, Ribli, D, Bagoly, Z, Csabai, I, Horvath, I & Balzs, LG 2017, 'New results in applying the machine learning to GRB redshift estimation', Proceedings of Science, vol. 2017-October.
Racz II, Ribli D, Bagoly Z, Csabai I, Horvath I, Balzs LG. New results in applying the machine learning to GRB redshift estimation. Proceedings of Science. 2017 Jan 1;2017-October.
Racz, Istvn I. ; Ribli, Dezso ; Bagoly, Z. ; Csabai, Istvn ; Horvath, Istvn ; Balzs, Lajos G. / New results in applying the machine learning to GRB redshift estimation. In: Proceedings of Science. 2017 ; Vol. 2017-October.
@article{215fb55058244f7387767a211ee9524f,
title = "New results in applying the machine learning to GRB redshift estimation",
abstract = "Gamma-ray bursts (GRBs) are the most energetic transients in the far Universe. Several thousands of GRBs have been observed so far but we could measure the distance of only a few hundreds. We studied the parameters of GRBs with available spectroscopic redshift in order to be able to estimate the redshift of those GRBs without a measured one. To calculate their distances we applied two machine-learning estimator methods: random forest regressor and XGBoost. For the process we used selected gamma, x-ray and ultraviolet parameters from the Swift GRB catalog, which contains the measured spectroscopic redshift of 328 GRBs. We found a significantly higher correlation between the measured and estimated redshift, we have improved the correlation in multiple steps from 0.57 (published by Ukwatta et al., 2016) to 0.67. It seems that both the random forest and the XGBoost methods give similarly high correlation. For further improvements additional redshift measurements are required.",
author = "Racz, {Istvn I.} and Dezso Ribli and Z. Bagoly and Istvn Csabai and Istvn Horvath and Balzs, {Lajos G.}",
year = "2017",
month = "1",
day = "1",
language = "English",
volume = "2017-October",
journal = "Proceedings of Science",
issn = "1824-8039",
publisher = "Sissa Medialab Srl",

}

TY - JOUR

T1 - New results in applying the machine learning to GRB redshift estimation

AU - Racz, Istvn I.

AU - Ribli, Dezso

AU - Bagoly, Z.

AU - Csabai, Istvn

AU - Horvath, Istvn

AU - Balzs, Lajos G.

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Gamma-ray bursts (GRBs) are the most energetic transients in the far Universe. Several thousands of GRBs have been observed so far but we could measure the distance of only a few hundreds. We studied the parameters of GRBs with available spectroscopic redshift in order to be able to estimate the redshift of those GRBs without a measured one. To calculate their distances we applied two machine-learning estimator methods: random forest regressor and XGBoost. For the process we used selected gamma, x-ray and ultraviolet parameters from the Swift GRB catalog, which contains the measured spectroscopic redshift of 328 GRBs. We found a significantly higher correlation between the measured and estimated redshift, we have improved the correlation in multiple steps from 0.57 (published by Ukwatta et al., 2016) to 0.67. It seems that both the random forest and the XGBoost methods give similarly high correlation. For further improvements additional redshift measurements are required.

AB - Gamma-ray bursts (GRBs) are the most energetic transients in the far Universe. Several thousands of GRBs have been observed so far but we could measure the distance of only a few hundreds. We studied the parameters of GRBs with available spectroscopic redshift in order to be able to estimate the redshift of those GRBs without a measured one. To calculate their distances we applied two machine-learning estimator methods: random forest regressor and XGBoost. For the process we used selected gamma, x-ray and ultraviolet parameters from the Swift GRB catalog, which contains the measured spectroscopic redshift of 328 GRBs. We found a significantly higher correlation between the measured and estimated redshift, we have improved the correlation in multiple steps from 0.57 (published by Ukwatta et al., 2016) to 0.67. It seems that both the random forest and the XGBoost methods give similarly high correlation. For further improvements additional redshift measurements are required.

UR - http://www.scopus.com/inward/record.url?scp=85041077334&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041077334&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:85041077334

VL - 2017-October

JO - Proceedings of Science

JF - Proceedings of Science

SN - 1824-8039

ER -