AUTOMATIC CLASSIFICATION OF KEPLER PLANETARY TRANSIT CANDIDATES

By S. McCauliff, Jon M. Jenkins, J. Catanzarite, C. Burke, Jeff Coughlin, Joseph D. Twicken, P. Tenenbaum, S. Seader, Jie Li, M. Cote. 2014

71 citations recorded by the source corpus.

In the first three years of operation, the Kepler mission found 3697 planet candidates (PCs) from a set of 18,406 transit-like features detected on more than 200,000 distinct stars. Vetting candidate signals manually by inspecting light curves and other diagnostic information is a labor intensive effort. Additionally, this classification methodology does not yield any information about the quality of PCs; all candidates are as credible as any other. The torrent of exoplanet discoveries will continue after Kepler, because a number of exoplanet surveys will have an even broader search area. This paper presents the application of machine-learning techniques to the classification of the exoplanet transit-like signals present in the Kepler light curve data. Transit-like detections are transformed into a uniform set of real-numbered attributes, the most important of which are described in this paper. Each of the known transit-like detections is assigned a class of PC; astrophysical false positive; or systematic, instrumental noise. We use a random forest algorithm to learn the mapping from attributes to classes on this training set. The random forest algorithm has been used previously to classify variable stars; this is the first time it has been used for exoplanet classification. We are able to achieve an overall error rate of 5.85% and an error rate for classifying exoplanets candidates of 2.81%.

Subfields: Astronomy and Astrophysics, Instrumentation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Astronomy and Astrophysics

In Instrumentation

Citations in the corpus

Identifying Exoplanets with Deep Learning: A Five-planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90 (2017, methodology): “Autovetter The Autovetter (McCauliff et al. 2015; Catanzarite 2015) is a machine learning system for classifying Kepler TCEs into 3 classes (planet candidates, astrophysical false positives and non-transiting phenomena).”
THE KEPLER FOLLOW-UP OBSERVATION PROGRAM. I. A CATALOG OF COMPANIONS TO KEPLER STARS FROM HIGH-RESOLUTION IMAGING (2016, background): “…process addressed in many publications (Fressin et al. 2013; Coughlin et al. 2014; Seader et al. 2015; Mullally et al. 2015; Desert et al. 2015; McCauliff et al. 2015; Santerne et al. 2016; Morton et al. 2016), with current estimates for false positive rates ranging from ∼ 10% for small…”
The California-Kepler Survey. I. High-resolution Spectroscopy of 1305 Stars Hosting Kepler Transiting Planets (2017, unknown): “…Gilliland et al. 2010; Stumpe et al. 2012; Smith et al. 2012, 2016; Batalha et al. 2010a,b; Torres et al. 2011; Bryson et al. 2013; Christiansen et al. 2012, 2013, 2015, 2016; Thompson et al. 2015; McCauliff et al. 2015; Tenenbaum et al. 2013, 2014; Twicken et al. 2016; Kinemuchi et al. 2012).”
TERRESTRIAL PLANET OCCURRENCE RATES FOR THE KEPLER GK DWARF SAMPLE (2015, methodology): “…Candidate Sample Characterizing a detection by the Kepler pipeline as a bona fide member of the Kepler planet candidate sample has increasingly relied upon an automated classification procedure (McCauliff et al. 2014; Rowe et al. 2015; Mullally et al. 2015; Coughlin et al. 2015; Rowe et al. 2015).”