Presentation #107.12 in the session Stellar/Compact Objects - Poster Session.
Globular clusters (GCs) are dense stellar environments that are rich with compact objects. They are also important testbeds for modeling stellar evolution in binaries. Over the last two decades, Chandra X-ray Observatory (CXO) discovered large populations of active binaries (ABs), cataclysmic variables (CVs), millisecond pulsars (MSPs), and low-mass X-ray binaries (LMXBs) in over 60 GCs. This was possible due to the unprecedented subarcsecond angular resolution of CXO, enabling the correct identification of optical counterparts from HST. However, the crowded environment still makes assigning the correct multiwavelength counterpart challenging, and over one thousand CXO sources remain unclassified. We cross-matched the Chandra Source Catalog (CSC) to the HST UV Globular Cluster Survey (HUGS) and created a training dataset of ~270 reliably classified GC X-ray sources (from the aforementioned classes). We use this training dataset to train a supervised machine-learning algorithm and classify these unclassified sources. For each CSC source we consider all likely optical associations to obtain a probabilistic classification. We present the training dataset, describe the machine learning classification pipeline, and report the initial results. Having both CXO and multi-band HST observations is crucial for the success of this synergetic approach.