SIDRA: A blind algorithm for signal detection in photometric surveys

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

We present the Signal Detection using Random-Forest Algorithm (SIDRA). SIDRA is a detection and classification algorithm based on the Machine Learning technique (Random Forest). The goal of this paper is to show the power of SIDRA for quick and accurate signal detection and classification.We first diagnose the power of the method with simulated light curves and try it on a subset of the Kepler space mission catalogue.We use five classes of simulated light curves (CONSTANT, TRANSIT, VARIABLE, MLENS and EB for constant light curves, transiting exoplanet, variable, microlensing events and eclipsing binaries, respectively) to analyse the power of the method. The algorithm uses four features in order to classify the light curves. The training sample contains 5000 light curves (1000 from each class) and 50 000 random light curves for testing. The total SIDRA success ratio is ≥90 per cent. Furthermore, the success ratio reaches 95-100 per cent for the CONSTANT, VARIABLE, EB and MLENS classes and 92 per cent for the TRANSIT class with a decision probability of 60 per cent. Because the TRANSIT class is the one which fails the most, we run a simultaneous fit using SIDRA and a Box Least Square (BLS)-based algorithm for searching for transiting exoplanets. As a result, our algorithm detects 7.5 per cent more planets than a classic BLS algorithm, with better results for lower signal-to-noise light curves. SIDRA succeeds to catch 98 per cent of the planet candidates in the Kepler sample and fails for 7 per cent of the false alarms subset. SIDRA promises to be useful for developing a detection algorithm and/or classifier for large photometric surveys such as TESS and PLATO exoplanet future space missions.

Original languageEnglish
Pages (from-to)626-633
Number of pages8
JournalMonthly Notices of the Royal Astronomical Society
Volume455
Issue number1
DOIs
Publication statusPublished - 2016

Fingerprint

signal detection
light curve
extrasolar planets
space missions
detection
set theory
boxes
planets
planet
machine learning
false alarms
transit
classifiers
catalogs
education

Keywords

  • Detection
  • Fundamental parameters
  • Photometric
  • Planetary systems
  • Planets and satellites
  • Planets and satellites
  • Techniques

ASJC Scopus subject areas

  • Space and Planetary Science
  • Astronomy and Astrophysics

Cite this

SIDRA : A blind algorithm for signal detection in photometric surveys. / Mislis, Dimitrios; Bachelet, E.; Al-Subai, Khalid; Bramich, D. M.; Parley, N.

In: Monthly Notices of the Royal Astronomical Society, Vol. 455, No. 1, 2016, p. 626-633.

Research output: Contribution to journalArticle

@article{73ee192e95dc4e939c251c7fd60d80b7,
title = "SIDRA: A blind algorithm for signal detection in photometric surveys",
abstract = "We present the Signal Detection using Random-Forest Algorithm (SIDRA). SIDRA is a detection and classification algorithm based on the Machine Learning technique (Random Forest). The goal of this paper is to show the power of SIDRA for quick and accurate signal detection and classification.We first diagnose the power of the method with simulated light curves and try it on a subset of the Kepler space mission catalogue.We use five classes of simulated light curves (CONSTANT, TRANSIT, VARIABLE, MLENS and EB for constant light curves, transiting exoplanet, variable, microlensing events and eclipsing binaries, respectively) to analyse the power of the method. The algorithm uses four features in order to classify the light curves. The training sample contains 5000 light curves (1000 from each class) and 50 000 random light curves for testing. The total SIDRA success ratio is ≥90 per cent. Furthermore, the success ratio reaches 95-100 per cent for the CONSTANT, VARIABLE, EB and MLENS classes and 92 per cent for the TRANSIT class with a decision probability of 60 per cent. Because the TRANSIT class is the one which fails the most, we run a simultaneous fit using SIDRA and a Box Least Square (BLS)-based algorithm for searching for transiting exoplanets. As a result, our algorithm detects 7.5 per cent more planets than a classic BLS algorithm, with better results for lower signal-to-noise light curves. SIDRA succeeds to catch 98 per cent of the planet candidates in the Kepler sample and fails for 7 per cent of the false alarms subset. SIDRA promises to be useful for developing a detection algorithm and/or classifier for large photometric surveys such as TESS and PLATO exoplanet future space missions.",
keywords = "Detection, Fundamental parameters, Photometric, Planetary systems, Planets and satellites, Planets and satellites, Techniques",
author = "Dimitrios Mislis and E. Bachelet and Khalid Al-Subai and Bramich, {D. M.} and N. Parley",
year = "2016",
doi = "10.1093/mnras/stv2333",
language = "English",
volume = "455",
pages = "626--633",
journal = "Monthly Notices of the Royal Astronomical Society",
issn = "0035-8711",
publisher = "Oxford University Press",
number = "1",

}

TY - JOUR

T1 - SIDRA

T2 - A blind algorithm for signal detection in photometric surveys

AU - Mislis, Dimitrios

AU - Bachelet, E.

AU - Al-Subai, Khalid

AU - Bramich, D. M.

AU - Parley, N.

PY - 2016

Y1 - 2016

N2 - We present the Signal Detection using Random-Forest Algorithm (SIDRA). SIDRA is a detection and classification algorithm based on the Machine Learning technique (Random Forest). The goal of this paper is to show the power of SIDRA for quick and accurate signal detection and classification.We first diagnose the power of the method with simulated light curves and try it on a subset of the Kepler space mission catalogue.We use five classes of simulated light curves (CONSTANT, TRANSIT, VARIABLE, MLENS and EB for constant light curves, transiting exoplanet, variable, microlensing events and eclipsing binaries, respectively) to analyse the power of the method. The algorithm uses four features in order to classify the light curves. The training sample contains 5000 light curves (1000 from each class) and 50 000 random light curves for testing. The total SIDRA success ratio is ≥90 per cent. Furthermore, the success ratio reaches 95-100 per cent for the CONSTANT, VARIABLE, EB and MLENS classes and 92 per cent for the TRANSIT class with a decision probability of 60 per cent. Because the TRANSIT class is the one which fails the most, we run a simultaneous fit using SIDRA and a Box Least Square (BLS)-based algorithm for searching for transiting exoplanets. As a result, our algorithm detects 7.5 per cent more planets than a classic BLS algorithm, with better results for lower signal-to-noise light curves. SIDRA succeeds to catch 98 per cent of the planet candidates in the Kepler sample and fails for 7 per cent of the false alarms subset. SIDRA promises to be useful for developing a detection algorithm and/or classifier for large photometric surveys such as TESS and PLATO exoplanet future space missions.

AB - We present the Signal Detection using Random-Forest Algorithm (SIDRA). SIDRA is a detection and classification algorithm based on the Machine Learning technique (Random Forest). The goal of this paper is to show the power of SIDRA for quick and accurate signal detection and classification.We first diagnose the power of the method with simulated light curves and try it on a subset of the Kepler space mission catalogue.We use five classes of simulated light curves (CONSTANT, TRANSIT, VARIABLE, MLENS and EB for constant light curves, transiting exoplanet, variable, microlensing events and eclipsing binaries, respectively) to analyse the power of the method. The algorithm uses four features in order to classify the light curves. The training sample contains 5000 light curves (1000 from each class) and 50 000 random light curves for testing. The total SIDRA success ratio is ≥90 per cent. Furthermore, the success ratio reaches 95-100 per cent for the CONSTANT, VARIABLE, EB and MLENS classes and 92 per cent for the TRANSIT class with a decision probability of 60 per cent. Because the TRANSIT class is the one which fails the most, we run a simultaneous fit using SIDRA and a Box Least Square (BLS)-based algorithm for searching for transiting exoplanets. As a result, our algorithm detects 7.5 per cent more planets than a classic BLS algorithm, with better results for lower signal-to-noise light curves. SIDRA succeeds to catch 98 per cent of the planet candidates in the Kepler sample and fails for 7 per cent of the false alarms subset. SIDRA promises to be useful for developing a detection algorithm and/or classifier for large photometric surveys such as TESS and PLATO exoplanet future space missions.

KW - Detection

KW - Fundamental parameters

KW - Photometric

KW - Planetary systems

KW - Planets and satellites

KW - Planets and satellites

KW - Techniques

UR - http://www.scopus.com/inward/record.url?scp=84958292246&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84958292246&partnerID=8YFLogxK

U2 - 10.1093/mnras/stv2333

DO - 10.1093/mnras/stv2333

M3 - Article

AN - SCOPUS:84958292246

VL - 455

SP - 626

EP - 633

JO - Monthly Notices of the Royal Astronomical Society

JF - Monthly Notices of the Royal Astronomical Society

SN - 0035-8711

IS - 1

ER -