Searching for compact objects within X-ray catalogs using Machine Learning
Abstract
Modern X-ray observatories (e.g., Chandra and XMM-Newton) have detected large numbers of Galactic sources serendipitously. The X-ray properties of these sources are extracted and then placed into catalogs, where they remain primarily unstudied. Novel discoveries and rare source classes are certain to be revealed by studying these rich datasets. However, X-ray data alone is often not enough to classify these sources, especially the faint sources whose population dominates these catalogs. Therefore, additional multiwavelength data must be used. MUWCLASS is a multiwavelength machine-learning pipeline that we have developed to classify X-ray sources. The pipeline relies on an extensive multiwavelength training dataset to carry out this task. In this talk, I will describe the training dataset, pipeline, and validation procedures used to construct MUWCLASS. I will then discuss the results of the classification of X-ray sources in multiple environments, including unidentified TeV sources, stellar clusters, and SNRs.
- Publication:
-
American Astronomical Society Meeting Abstracts #233
- Pub Date:
- January 2019
- Bibcode:
- 2019AAS...23311005H