Automatic Classification of Variable Stars in Catalogs with missing data

Abstract

We present an automatic classification method for astronomical catalogs with missing data. We use Bayesian networks, a probabilistic graphical model, that allows us to perform inference to pre- dict missing values given observed data and dependency relationships between variables. To learn a Bayesian network from incomplete data, we use an iterative algorithm that utilises sampling methods and expectation maximization to estimate the distributions and probabilistic dependencies of variables from data with missing values. To test our model we use three catalogs with missing data (SAGE, 2MASS and UBVI) and one complete catalog (MACHO). We examine how classification accuracy changes when information from missing data catalogs is included, how our method compares to traditional missing data approaches and at what computational cost. Integrating these catalogs with missing data we find that classification of variable objects improves by few percent and by 15% for quasar detection while keeping the computational cost the same.

References in corpus (2)

• Automated supervised classification of variable stars I. Methodology
• An improved quasar detection method in EROS-2 and MACHO LMC datasets

Cited by in corpus (32)

• The Automatic Learning for the Rapid Classification of Events (ALeRCE) Alert Broker
• Probabilistic Random Forest: A machine learning algorithm for noisy datasets
• Computational Intelligence Challenges and Applications on Large-Scale Astronomical Time Series Databases
• Scalable End-to-end Recurrent Neural Network for Variable star classification
• Supervised detection of anomalous light-curves in massive astronomical catalogs
• Deep multi-survey classification of variable stars
• Deep Learning for Image Sequence Classification of Astronomical Events
• Unsupervised machine learning for transient discovery in Deeper, Wider, Faster light curves
• The VVV Templates Project. Towards an Automated Classification of VVV Light-Curves. I. Building a database of stellar variability in the near-infrared
• Clustering Based Feature Learning on Variable Stars
• Unsupervised Classification of Variable Stars
• Meta Classification for Variable Stars
• Photometric Classification of quasars from RCS-2 using Random Forest
• The High Cadence Transient Survey (HITS): Compilation and characterization of light-curve catalogs
• Automatic Survey-Invariant Variable Star Classification
• The VVV Infrared Variability Catalog (VIVA-I)
• Uncertain classification of Variable Stars: handling observational GAPS and noise
• Generation of a Supervised Classification Algorithm for Time-Series Variable Stars with an Application to the LINEAR Dataset
• Streaming Classification of Variable Stars
• Semi-Supervised Classification and Clustering Analysis for Variable Stars
• On the Use of Logistic Regression for stellar classification. An application to colour-colour diagrams
• Selection of Burst-like Transients and Stochastic Variables Using Multi-Band Image Differencing in the Pan-STARRS1 Medium-Deep Survey
• A Detection Metric Designed for O'Connell Effect Eclipsing Binaries
• ASTROMLSKIT: A New Statistical Machine Learning Toolkit: A Platform for Data Analytics in Astronomy
• A Fast Approximate Approach to Microlensing Survey Analysis
• Variable Star Signature Classification using Slotted Symbolic Markov Modeling
• Classifying High-cadence Microlensing Light Curves I; Defining Features
• On the use of variability time-scales as an early classifier of radio transients and variables
• Science Fiction Media Representations of Exoplanets: Portrayals of Changing Astronomical Discoveries
• An Information Theory Approach on Deciding Spectroscopic Follow Ups
• An Algorithm for the Visualization of Relevant Patterns in Astronomical Light Curves
• Advanced Astroinformatics for Variable Star Classification