This study investigates the synthesis, characterization, and dielectric spectroscopy of epoxy composites reinforced with ZnO and TiO₂ nanoparticles. The epoxy resin was crosslinked using Bisphenol-A and an amine hardener, forming a three-dimensional structure. Dielectric spectroscopic measurements were conducted over a wide frequency range (20 Hz to 20 GHz) to evaluate the impact of ZnO and TiO₂ nanoparticles on the dielectric properties of the epoxy matrix. The research provides a detailed analysis of how these nanoparticles influence the dielectric behavior of the composite material. Regression analysis, particularly XGBoost, was employed to predict the dielectric constant of the composites, offering a more efficient method than traditional experimental approaches. The results show that XGBoost can reduce both time and resource requirements by up to 60%. The study highlights the importance of understanding the dielectric properties of ZnO and TiO₂ nanoparticle-doped composites for applications in capacitors, insulators, microwave and RF devices, energy storage, and sensors. Previous studies have shown that the addition of ZnO nanoparticles can improve mechanical properties such as bending strength and flexural modulus, while also affecting the dielectric constant. The integration of machine learning techniques into materials characterization offers a promising approach for predicting material properties efficiently. This study aims to comprehensively investigate the dielectric properties of ZnO and TiO₂ nanoparticle-doped epoxy composites and to develop robust prediction models using machine learning algorithms. The findings emphasize the potential of these composites in various electronic applications, particularly in the radio and microwave frequency range.This study investigates the synthesis, characterization, and dielectric spectroscopy of epoxy composites reinforced with ZnO and TiO₂ nanoparticles. The epoxy resin was crosslinked using Bisphenol-A and an amine hardener, forming a three-dimensional structure. Dielectric spectroscopic measurements were conducted over a wide frequency range (20 Hz to 20 GHz) to evaluate the impact of ZnO and TiO₂ nanoparticles on the dielectric properties of the epoxy matrix. The research provides a detailed analysis of how these nanoparticles influence the dielectric behavior of the composite material. Regression analysis, particularly XGBoost, was employed to predict the dielectric constant of the composites, offering a more efficient method than traditional experimental approaches. The results show that XGBoost can reduce both time and resource requirements by up to 60%. The study highlights the importance of understanding the dielectric properties of ZnO and TiO₂ nanoparticle-doped composites for applications in capacitors, insulators, microwave and RF devices, energy storage, and sensors. Previous studies have shown that the addition of ZnO nanoparticles can improve mechanical properties such as bending strength and flexural modulus, while also affecting the dielectric constant. The integration of machine learning techniques into materials characterization offers a promising approach for predicting material properties efficiently. This study aims to comprehensively investigate the dielectric properties of ZnO and TiO₂ nanoparticle-doped epoxy composites and to develop robust prediction models using machine learning algorithms. The findings emphasize the potential of these composites in various electronic applications, particularly in the radio and microwave frequency range.