Density Based Outlier Scoring on Kepler Data

Abstract

In the present era of large scale surveys, big data presents new challenges to the discovery process for anomalous data. Such data can be indicative of systematic errors, extreme (or rare) forms of known phenomena, or most interestingly, truly novel phenomena which exhibit as-of-yet unobserved behaviors. In this work we present an outlier scoring methodology to identify and characterize the most promising unusual sources to facilitate discoveries of such anomalous data. We have developed a data mining method based on k-Nearest Neighbor distance in feature space to efficiently identify the most anomalous lightcurves. We test variations of this method including using principal components of the feature space, removing select features, the effect of the choice of k, and scoring to subset samples. We evaluate the peformance of our scoring on known object classes and find that our scoring consistently scores rare (<1000) object classes higher than common classes. We have applied scoring to all long cadence lightcurves of quarters 1 to 17 of Kepler's prime mission and present outlier scores for all 2.8 million lightcurves for the roughly 200k objects.

14 pages, 11 figures, plus appendices, 4 supplemental tables hosted online at https://doi.org/10.7910/DVN/H5QXUL

References in corpus (34)

• Scikit-learn: Machine Learning in Python
• Gaia Data Release 2. Summary of the contents and survey properties
• The Transiting Exoplanet Survey Satellite
• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• The SIMBAD astronomical database
• Gaia Data Release 1. Summary of the astrometric, photometric, and survey properties
• Galaxy Zoo Green Peas: Discovery of A Class of Compact Extremely Star-Forming Galaxies
• Kepler Eclipsing Binary Stars. VII. The Catalog of Eclipsing Binaries Found in the Entire Kepler Data-Set
• Data Mining and Machine Learning in Astronomy
• The Kepler Catalog of Stellar Flares
• On Machine-Learned Classification of Variable Stars with Sparse and Noisy Time-Series Data
• Selection, Prioritization, and Characteristics of Kepler Target Stars
• Galaxy Zoo : 'Hanny's Voorwerp', a quasar light echo?
• Planet Hunters X. KIC 8462852 - Where's the Flux?
• Kepler Mission Stellar and Instrument Noise Properties
• Planetary Candidates Observed by Kepler. VII. The First Fully Uniform Catalog Based on The Entire 48 Month Dataset (Q1-Q17 DR24)
• A Class of Eccentric Binaries with Dynamic Tidal Distortions Discovered with Kepler
• A new method for unveiling Open Clusters in Gaia: new nearby Open Clusters confirmed by DR2
• Kepler Eclipsing Binary Stars. III. Classification of Kepler Eclipsing Binary Light Curves with Locally Linear Embedding
• Results from the Supernova Photometric Classification Challenge
• Automating Discovery and Classification of Transients and Variable Stars in the Synoptic Survey Era
• Machine Learning in Astronomy: a practical overview
• The weirdest SDSS galaxies: results from an outlier detection algorithm
• Global stellar variability study in the field-of-view of the Kepler satellite
• The Ĝ Infrared Search for Extraterrestrial Civilizations with Large Energy Supplies. II. Framework, Strategy, and First Result
• Finding outlier light-curves in catalogs of periodic variable stars
• Kepler Mission Stellar and Instrument Noise Properties Revisited
• The First Post-Kepler Brightness Dips of KIC 8462852
• Unusual quasars from the Sloan Digital Sky Survey selected by means of Kohonen self-organising maps
• Systematic Serendipity: A Test of Unsupervised Machine Learning as a Method for Anomaly Detection
• The Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC): Data set
• Machine learning in APOGEE: Unsupervised spectral classification with $K$-means
• Supernova Photometric Classification Challenge
• Identifying complex sources in large astronomical data using a coarse-grained complexity measure

Cited by in corpus (10)

• Anomaly detection in the Zwicky Transient Facility DR3
• Classifying Kepler light curves for 12,000 A and F stars using supervised feature-based machine learning
• A method for finding anomalous astronomical light curves and their analogs
• Exploring X-ray variability with unsupervised machine learning I. Self-organizing maps applied to XMM-Newton data
• Identifying Light-curve Signals with a Deep Learning Based Object Detection Algorithm. II. A General Light Curve Classification Framework
• Identifying anomalous radio sources in the EMU Pilot Survey using a complexity-based approach
• SETI in 2020
• Automated Discovery of Anomalous Features in Ultra-Large Planetary Remote Sensing Datasets using Variational Autoencoders
• Interpreting anomaly detection of SDSS spectra
• Searching for technosignatures in exoplanetary systems with current and future missions