CaltechAUTHORS
  A Caltech Library Service

An analysis of feature relevance in the classification of astronomical transients with machine learning methods

D'Isanto, A. and Cavuoti, S. and Brescia, M. and Donalek, C. and Longo, G. and Riccio, G. and Djorgovski, S. G. (2016) An analysis of feature relevance in the classification of astronomical transients with machine learning methods. Monthly Notices of the Royal Astronomical Society, 457 (3). pp. 3119-3132. ISSN 0035-8711. http://resolver.caltech.edu/CaltechAUTHORS:20160509-092928454

[img] PDF - Published Version
See Usage Policy.

2136Kb
[img] PDF - Submitted Version
See Usage Policy.

2623Kb

Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:20160509-092928454

Abstract

The exploitation of present and future synoptic (multiband and multi-epoch) surveys requires an extensive use of automatic methods for data processing and data interpretation. In this work, using data extracted from the Catalina Real Time Transient Survey (CRTS), we investigate the classification performance of some well tested methods: Random Forest, MultiLayer Perceptron with Quasi Newton Algorithm and K-Nearest Neighbours, paying special attention to the feature selection phase. In order to do so, several classification experiments were performed. Namely: identification of cataclysmic variables, separation between galactic and extragalactic objects and identification of supernovae.


Item Type:Article
Related URLs:
URLURL TypeDescription
http://dx.doi.org/10.1093/mnras/stw157DOIArticle
https://mnras.oxfordjournals.org/content/457/3/3119PublisherArticle
http://arxiv.org/abs/1601.03931arXivDiscussion Paper
ORCID:
AuthorORCID
Djorgovski, S. G.0000-0002-0603-3087
Additional Information:© 2016 The Authors Published by Oxford University Press on behalf of the Royal Astronomical Society. Accepted 2016 January 15. Received 2016 January 15. In original form 2015 October 30. First published online February 20, 2016. The authors wish to thank the anonymous referee for the very helpful comments and suggestions which contributed to optimize the manuscript. The authors wish to acknowledge financial support from the Italian Ministry of University and Research through the grant PRIN-MIUR ‘Cosmology with Euclid’ and from the Keck Institute for Space Studies who sponsored the working group on TDA. MB and SC acknowledge financial contribution from the agreement ASI/INAF I/023/12/1. Authors also wish to thank K. Polsterer for useful discussions. This work made use of the CRTS public archive data, of the CTSCS service and of the DAMEWARE infrastructure. CD and SGD acknowledge a partial support from the NSF grants AST-1413600 and AST-1313422. We are thankful to A. J. Drake, A. A. Mahabal, and M. J. Graham for their key contributions in the CRTS project.
Group:Keck Institute for Space Studies
Funders:
Funding AgencyGrant Number
Ministero dell'Istruzione, dell'Università e della Ricerca (MIUR)UNSPECIFIED
Keck Institute for Space Studies UNSPECIFIED
Agenzia Spaziale Italiana (ASI)I/023/12/1
NSFAST-1413600
NSFAST-1313422
Istituto Nazionale di Astrofisica (INAF)UNSPECIFIED
Subject Keywords:methods: data analysis novae, cataclysmic variables supernovae: general stars: variables: general stars: variables: RR Lyrae
Record Number:CaltechAUTHORS:20160509-092928454
Persistent URL:http://resolver.caltech.edu/CaltechAUTHORS:20160509-092928454
Official Citation:A. D'Isanto, S. Cavuoti, M. Brescia, C. Donalek, G. Longo, G. Riccio, and S. G. Djorgovski An analysis of feature relevance in the classification of astronomical transients with machine learning methods MNRAS (April 11, 2016) Vol. 457 3119-3132 doi:10.1093/mnras/stw157 First published online February 20, 2016
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:66731
Collection:CaltechAUTHORS
Deposited By: Ruth Sustaita
Deposited On:09 May 2016 17:04
Last Modified:09 May 2016 17:04

Repository Staff Only: item control page