This study investigates the synthesis, characterization, and dielectric properties of epoxy composites reinforced with TiO₂ and ZnO nanoparticles. The research focuses on the impact of these nanoparticles on the dielectric behavior of the composites, aiming to enhance their performance in electronic applications. A three-dimensional crosslinked structure was formed using Bisphenol-A epoxy resin and an amine hardener. Dielectric spectroscopic assessments were conducted over a frequency range from 20 Hz to 2 GHz, with a detailed analysis of the influence of ZnO and TiO₂ nanoparticles on the pristine epoxy resin. The study also employs Regression analysis, particularly XGBoost, to predict the dielectric constant, which can significantly reduce the time and resource requirements for such analyses. The findings highlight the potential of these nanocomposites in various electronic devices, including capacitors, insulators, microwave and RF devices, and sensor technologies. The integration of machine learning techniques into materials characterization is discussed as a promising approach to overcome the challenges of traditional experimental methods.This study investigates the synthesis, characterization, and dielectric properties of epoxy composites reinforced with TiO₂ and ZnO nanoparticles. The research focuses on the impact of these nanoparticles on the dielectric behavior of the composites, aiming to enhance their performance in electronic applications. A three-dimensional crosslinked structure was formed using Bisphenol-A epoxy resin and an amine hardener. Dielectric spectroscopic assessments were conducted over a frequency range from 20 Hz to 2 GHz, with a detailed analysis of the influence of ZnO and TiO₂ nanoparticles on the pristine epoxy resin. The study also employs Regression analysis, particularly XGBoost, to predict the dielectric constant, which can significantly reduce the time and resource requirements for such analyses. The findings highlight the potential of these nanocomposites in various electronic devices, including capacitors, insulators, microwave and RF devices, and sensor technologies. The integration of machine learning techniques into materials characterization is discussed as a promising approach to overcome the challenges of traditional experimental methods.