Construction of a Calibrated Probabilistic Classification Catalog: Application to 50k Variable Sources in the All-Sky Automated Survey

TL;DR

This paper presents a machine learning-based probabilistic classification catalog for 50,124 variable sources from the ASAS survey, emphasizing accurate, calibrated class probabilities and anomaly detection to improve follow-up efficiency.

Contribution

It introduces a novel methodology for producing a calibrated, multi-class probabilistic catalog from survey data, including anomaly detection, and applies it to create the MACC for ASAS sources.

Findings

Achieves sub-20% classification error rate

Provides well-calibrated class posterior probabilities

Outperforms previous ASAS classification efforts

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.