Automatic Classification of Variable Stars in Catalogs with missing data
This work addresses the challenge of handling missing data in astronomical catalogs for researchers in astronomy and data science, representing an incremental advancement over traditional methods.
The authors tackled the problem of classifying variable stars in astronomical catalogs with missing data using Bayesian networks, achieving improvements of a few percent in variable object classification and 15% in quasar detection while maintaining computational cost.
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.