Construction of a Calibrated Probabilistic Classification Catalog: Application to 50k Variable Sources in the All-Sky Automated Survey
arXiv:1204.4180 · doi:10.1088/0067-0049/203/2/32
Abstract
With growing data volumes from synoptic surveys, astronomers must become more abstracted from the discovery and introspection processes. Given the scarcity of follow-up resources, there is a particularly sharp onus on the frameworks that replace these human roles to provide accurate and well-calibrated probabilistic classification catalogs. Such catalogs inform the subsequent follow-up, allowing consumers to optimize the selection of specific sources for further study and permitting rigorous treatment of purities and efficiencies for population studies. Here, we describe a process to produce a probabilistic classification catalog of variability with machine learning from a multi-epoch photometric survey. In addition to producing accurate classifications, we show how to estimate calibrated class probabilities, and motivate the importance of probability calibration. We also introduce a methodology for feature-based anomaly detection, which allows discovery of objects in the survey that do not fit within the predefined class taxonomy. Finally, we apply these methods to sources observed by the All Sky Automated Survey (ASAS), and unveil the Machine-learned ASAS Classification Catalog (MACC), which is a 28-class probabilistic classification catalog of 50,124 ASAS sources. We estimate that MACC achieves a sub-20% classification error rate, and demonstrate that the class posterior probabilities are reasonably calibrated. MACC classifications compare favorably to the classifications of several previous domain-specific ASAS papers and to the ASAS Catalog of Variable Stars, which had classified only 24% of those sources into one of 12 science classes. The MACC is publicly available at http://www.bigmacc.info.
arXiv comment56 pages, 15 figures, 8 tables, submitted. The Machine-learned ASAS Classification Catalog is available at http://www.bigmacc.info
References in corpus (20)
- MissForest - nonparametric missing value imputation for mixed-type data
- The SIMBAD astronomical database
- The generalised Lomb-Scargle periodogram. A new formalism for the floating-mean and Keplerian periodograms
- On Machine-Learned Classification of Variable Stars with Sparse and Noisy Time-Series Data
- Large Synoptic Survey Telescope: Overview
- Automated supervised classification of variable stars I. Methodology
- The Evolution of Massive Stars. I. Red Supergiants in the Magellanic Clouds
- Random forest automated supervised classification of Hipparcos periodic variable stars
- Yellow and Red Supergiants in the Large Magellanic Cloud
- QSO Selection Algorithm Using Time Variability and Machine Learning: Selection of 1,620 QSO Candidates from MACHO LMC Database
- Finding outlier light-curves in catalogs of periodic variable stars
- Active Learning to Overcome Sample Selection Bias: Application to Photometric Variable Star Classification
- Multiperiodic Galactic field RR Lyrae stars in the ASAS catalog
- Detecting Variability in Massive Astronomical Time-Series Data I: application of an infinite Gaussian mixture model
- Automated classification of periodic variable stars{Improved methodology for the automated classification of periodic variable stars}
- Discovery of Bright Galactic R Coronae Borealis and DY Persei Variables: Rare Gems Mined from ACVS
- Optimizing Automated Classification of Periodic Variable Stars in New Synoptic Surveys
- Parameter Estimation with BEAMS in the presence of biases and correlations
- Finding the most variable stars in the Orion Belt with the All Sky Automated Survey
- Variability type classification of multi-epoch surveys
Cited by in corpus (70)
- Understanding the Lomb-Scargle Periodogram
- The Catalina Surveys Periodic Variable Star Catalog
- The Automatic Learning for the Rapid Classification of Events (ALeRCE) Alert Broker
- Gaia Data Release 2: Variable stars in the colour-absolute magnitude diagram
- Alert Classification for the ALeRCE Broker System: The Light Curve Classifier
- A re-classification of Cepheids in the Gaia Data Release 2
- A recurrent neural network for classification of unevenly sampled variable stars
- VAST: An ASKAP Survey for Variables and Slow Transients
- Gaia Data Release 3: All-sky classification of 12.4 million variable sources into 25 classes
- K2 Variable Catalogue II: Machine Learning Classification of Variable Stars and Eclipsing Binaries in K2 Fields 0-4
- Catalogue and Properties of δ Scuti Stars in Binaries
- A Morphological Classification Model to Identify Unresolved PanSTARRS1 Sources: Application in the ZTF Real-Time Pipeline
- A census of symbiotic stars in the 2MASS, WISE and Gaia surveys
- A comparison of period finding algorithms
- A Package for the Automated Classification of Periodic Variable Stars
- The magnetic fields of intermediate-mass T Tauri stars : I. Magnetic detections and fundamental stellar parameters
- Using conditional entropy to identify periodicity
- Gaia Data Release 3: Cross-match of Gaia sources with variable objects from the literature
- Multiplicity of Galactic Cepheids and RR Lyrae stars from Gaia DR2 -- II. Resolved common proper motion pairs
- IRC-10414: a bow-shock-producing red supergiant star
- Supervised detection of anomalous light-curves in massive astronomical catalogs
- Transit Shapes and Self Organising Maps as a Tool for Ranking Planetary Candidates: Application to Kepler and K2
- A machine learned classifier for RR Lyrae in the VVV survey
- Gaia Data Release 2: All-sky classification of high-amplitude pulsating stars
- On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars
- A search for photometric variability in magnetic chemically peculiar stars using ASAS-3 data
- A Machine Learning Method to Infer Fundamental Stellar Parameters from Photometric Light Curves
- New Photometrically Variable Magnetic Chemically Peculiar Stars in the ASAS-3 Archive
- Tracing the Milky Way warp and spiral arms with classical Cepheids
- Automatic vetting of planet candidates from ground based surveys: Machine learning with NGTS
- Planets Around Low-Mass Stars (PALMS). V. Age-Dating Low-Mass Companions to Members and Interlopers of Young Moving Groups
- MKT J170456.2-482100: the first transient discovered by MeerKAT
- Into the Darkness: Classical and Type II Cepheids in the Zona Galactica Incognita
- Discovery of New Dipper Stars with K2: A Window into the Inner Disk Region of T Tauri Stars
- The Variable Sky of Deep Synoptic Surveys
- A Mid-infrared Study of RR Lyrae Stars with the WISE All-Sky Data Release
- Variability, periodicity and contact binaries in WISE
- The Photometric LSST Astronomical Time-series Classification Challenge (PLAsTiCC): Selection of a performance metric for classification probabilities balancing diverse science goals
- The High Cadence Transient Survey (HITS): Compilation and characterization of light-curve catalogs
- Searching for dark-matter waves with PPTA and QUIJOTE pulsar polarimetry
- Magnetic, chemically peculiar (CP2) stars in the SuperWASP survey
- Preparing for advanced LIGO: A Star-Galaxy Separation Catalog for the Palomar Transient Factory
- VISTA variables in the Sagittarius dwarf spheroidal galaxy: pulsation- vs. dust-driven winds on the giant branches
- Four winters of photometry with ASTEP South at Dome C, Antarctica
- Variable classification in the LSST era: Exploring a model for quasi-periodic light curves
- Angular Momentum Evolution of Young Stars in the Nearby Scorpius-Centaurus OB Association
- Determination of pulsation periods and other parameters of 2875 stars classified as MIRA in the All Sky Automated Survey (ASAS)
- Generation of a Supervised Classification Algorithm for Time-Series Variable Stars with an Application to the LINEAR Dataset
- Near-Infrared Search for Fundamental-mode RR Lyrae Stars Toward the Inner Bulge by Deep Learning
- The EB Factory Project I. A Fast, Neural Net Based, General Purpose Light Curve Classifier Optimized for Eclipsing Binaries
- A Detection Metric Designed for O'Connell Effect Eclipsing Binaries
- Variable Star Classification Using Multi-View Metric Learning
- New Insights into Time Series Analysis III - Setting constraints on period analysis
- Confirmation of Monoperiodicity Above 20 Seconds for Two Blue Large-Amplitude Pulsators
- Variable Star Signature Classification using Slotted Symbolic Markov Modeling
- Recovering variable stars in large surveys: EA Algol-type class in the Catalina Survey
- Circumstellar CO J = 3-2 detected around the evolving metal-poor ([Fe/H] ~ -1.15 dex) AGB star RU Vulpeculae
- Exploring circumstellar effects on the lithium and calcium abundances in massive Galactic O-rich AGB stars
- Automatic Catalog of RRLyrae from 14 million VVV Light Curves: How far can we go with traditional machine-learning?
- A review of pulsating stars from the ASAS data
- YOUNG Star detrending for Transiting Exoplanet Recovery (YOUNGSTER) II: Using Self-Organising Maps to explore young star variability in Sectors 1-13 of TESS data
- TESS discovery of a super-Earth and two sub-Neptunes orbiting the bright, nearby, Sun-like star HD 22946
- HD 63021: Chromospheric Activity and Mass Transfer in a Close Binary
- Multiperiodic semiregular variable stars in the ASAS data base: A pilot study
- A Template-Based Approach to the Photometric Classification of SN 1991bg-like Supernovae in the SDSS-II Supernova Survey
- GRAPE: Genetic Routine for Astronomical Period Estimation
- A search for photometric variability towards the globular cluster M3 with the TWenty Inch Survey Telescope
- The First Multiband Photometric Light Curve Solutions of the V Gru Binary System from the Southern Hemisphere
- Search of nearby resolved neutron stars among optical sources
- BSN-VII: Photometric Light Curve Study of Three Cool, Large-Amplitude, Low Mass, W-Subtype Contact Binaries