Analysis of quantitative classification and properties of X-ray binary systems

X-ray astronomy is a field of research focusing on astronomical objects that emit X-ray radiation. The first interstellar X-ray source, Scorpius X-1, was discovered in 1962. Since then, hundreds of X-ray binaries (XRBs) have been discovered inside the Milky Way, with many unusual properties. These binary systems are currently classified as either high- or low-mass XRBs. However, it is useful to the field to have a classification system consisting of more properties than just mass to have a more descriptive classification of a stellar object. We aimed to analyze the patterns in several variables contributing to an XRB’s properties to find better-fit ways to classify these systems. We hypothesized that there would be a large difference in classes of XRBs across several variables, not just mass, but including the characteristics of a companion star. Utilizing several machine learning algorithms, we analyzed several variables describing an XRB. Dimensionality reduction was used to find correlations between the most important variables defining a star such as temperature, orbital period, metallicity, and redshift, and the classification groups each star falls under. We found that temperature is a quantitative parameter to classify these systems. There are three main groups of temperatures that XRBs fall into. These being stars with a metallicity <-2 and <100000K, those with a metallicity >-2 and <100000K, and those with a metallicity >-2 and >100000K. Characterizing XRBs are important to find star parameters that explain the chemical evolution of the galaxy and understanding of how binary parameters evolve toward detectable mergers.