Using Multivariate Imputation by Chained Equations to Predict Redshifts of Active Galactic Nuclei
(2022) In Frontiers in Astronomy and Space Sciences 9.- Abstract
Redshift measurement of active galactic nuclei (AGNs) remains a time-consuming and challenging task, as it requires follow up spectroscopic observations and detailed analysis. Hence, there exists an urgent requirement for alternative redshift estimation techniques. The use of machine learning (ML) for this purpose has been growing over the last few years, primarily due to the availability of large-scale galactic surveys. However, due to observational errors, a significant fraction of these data sets often have missing entries, rendering that fraction unusable for ML regression applications. In this study, we demonstrate the performance of an imputation technique called Multivariate Imputation by Chained Equations (MICE), which rectifies... (More)
Redshift measurement of active galactic nuclei (AGNs) remains a time-consuming and challenging task, as it requires follow up spectroscopic observations and detailed analysis. Hence, there exists an urgent requirement for alternative redshift estimation techniques. The use of machine learning (ML) for this purpose has been growing over the last few years, primarily due to the availability of large-scale galactic surveys. However, due to observational errors, a significant fraction of these data sets often have missing entries, rendering that fraction unusable for ML regression applications. In this study, we demonstrate the performance of an imputation technique called Multivariate Imputation by Chained Equations (MICE), which rectifies the issue of missing data entries by imputing them using the available information in the catalog. We use the Fermi-LAT Fourth Data Release Catalog (4LAC) and impute 24% of the catalog. Subsequently, we follow the methodology described in Dainotti et al. (ApJ, 2021, 920, 118) and create an ML model for estimating the redshift of 4LAC AGNs. We present results which highlight positive impact of MICE imputation technique on the machine learning models performance and obtained redshift estimation accuracy.
(Less)
- author
- Gibson, Spencer James ; Narendra, Aditya ; Dainotti, Maria Giovanna ; Bogdan, Malgorzata LU ; Pollo, Agnieszka ; Poliszczuk, Artem ; Rinaldi, Enrico and Liodakis, Ioannis
- organization
- publishing date
- 2022
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- AGNs, BLLs, FERMI 4LAC, FSRQs, imputation, machine learning regressors, MICE, redshift
- in
- Frontiers in Astronomy and Space Sciences
- volume
- 9
- article number
- 836215
- publisher
- Frontiers Media S. A.
- external identifiers
-
- scopus:85127149465
- ISSN
- 2296-987X
- DOI
- 10.3389/fspas.2022.836215
- language
- English
- LU publication?
- yes
- id
- e45f602e-3346-48d2-8a8f-a7b6f781951c
- date added to LUP
- 2022-06-02 11:20:09
- date last changed
- 2022-11-18 18:59:27
@article{e45f602e-3346-48d2-8a8f-a7b6f781951c,
abstract = {{<p>Redshift measurement of active galactic nuclei (AGNs) remains a time-consuming and challenging task, as it requires follow up spectroscopic observations and detailed analysis. Hence, there exists an urgent requirement for alternative redshift estimation techniques. The use of machine learning (ML) for this purpose has been growing over the last few years, primarily due to the availability of large-scale galactic surveys. However, due to observational errors, a significant fraction of these data sets often have missing entries, rendering that fraction unusable for ML regression applications. In this study, we demonstrate the performance of an imputation technique called Multivariate Imputation by Chained Equations (MICE), which rectifies the issue of missing data entries by imputing them using the available information in the catalog. We use the Fermi-LAT Fourth Data Release Catalog (4LAC) and impute 24% of the catalog. Subsequently, we follow the methodology described in Dainotti et al. (ApJ, 2021, 920, 118) and create an ML model for estimating the redshift of 4LAC AGNs. We present results which highlight positive impact of MICE imputation technique on the machine learning models performance and obtained redshift estimation accuracy.</p>}},
author = {{Gibson, Spencer James and Narendra, Aditya and Dainotti, Maria Giovanna and Bogdan, Malgorzata and Pollo, Agnieszka and Poliszczuk, Artem and Rinaldi, Enrico and Liodakis, Ioannis}},
issn = {{2296-987X}},
keywords = {{AGNs; BLLs; FERMI 4LAC; FSRQs; imputation; machine learning regressors; MICE; redshift}},
language = {{eng}},
publisher = {{Frontiers Media S. A.}},
series = {{Frontiers in Astronomy and Space Sciences}},
title = {{Using Multivariate Imputation by Chained Equations to Predict Redshifts of Active Galactic Nuclei}},
url = {{http://dx.doi.org/10.3389/fspas.2022.836215}},
doi = {{10.3389/fspas.2022.836215}},
volume = {{9}},
year = {{2022}},
}