Composite materials, renowned for their superior mechanical and thermal properties, are integral to various industries such as aerospace, automotive, and construction. The purpose of this study is to investigate the use of machine learning models to classify composite materials into three distinct classes: Carbon Nanotubes, Graphene, and Polymer Matrix. A dataset comprising material properties such as tensile strength, thermal conductivity, elastic modulus, density, and fiber volume fraction was utilized for training and evaluation. Among the evaluated models, XGBoost emerged as the best-performing model, achieving precision, recall, and F1-scores of 0.95, 0.93, and 0.94 for Carbon Nanotubes; 0.95, 0.94, and 0.94 for Graphene; and 0.92, 0.95, and 0.93 for Polymer Matrix, respectively. Random Forest and Multi-Layer Perceptron (MLP) also demonstrated strong performance, with metrics exceeding 0.93 for most classes. Results show the ability of advanced ensemble methods and neural networks to accurately classify composite materials, where dependency features have complex interactions and are not necessarily linear. By automating the classification process, this study advances a growing field of materials informatics.