X-ray diffraction (XRD) is an important tool for material science, allowing researchers to understand material microstructures and characterize material properties. It is especially useful for studying materials while they are subjected to extreme conditions like high temperatures and pressures. However, manual analysis of diffraction patterns is time-consuming and reliant on expert interpretation. Furthermore, manual analysis cannot accurately quantify solid and liquid phase fractions for a melting material, which contains a mix of both phases. Our project, under the guidance of Dr. Avinash M. Dongare (Materials Science) and Dr. Qian Yang (Computing), focused on developing machine learning models to automatically analyze XRD patterns and predict the fraction of the material in the solid and liquid phases. Using molecular dynamics data of melting copper from Dr. Dongare's GEMMS lab, we created a data processing pipeline to generate a large dataset of simulated XRD patterns. Then, we developed, trained and tested various machine learning models. Our best model achieved extremely high accuracy in predicting solid fractions, achieving an R² score over 0.99. Our XRD analysis approach can significantly reduce analysis time and provide precise quantitative results, enabling more efficient and accurate materials characterization workflows.