<p>Machine learning techniques are increasingly used in materials science and manufacturing to support material classification and selection. This study investigates the application of K-Means clustering, Decision Tree, and Support Vector Machine (SVM) algorithms to categorize engineering materials based on density and strength. Data for 43 materials were extracted from a strength–density chart in Materials Selection in Mechanical Design. K-Means clustering was first employed to group the unlabeled data into four classes, which were then used as labels for supervised learning. Decision Tree and SVM classifiers were trained and evaluated on this dataset to predict material categories. The Decision Tree achieved an accuracy of 100%, while the SVM achieved an accuracy of 81.4%, indicating that the Decision Tree provided more reliable classification for this dataset. This work highlights the potential of combining unsupervised and supervised learning methods for effective material selection and performance evaluation. This study demonstrates that a hybrid learning approach combining K-Means clustering with supervised classification enables accurate material categorisation using limited property data, with the Decision Tree model outperforming SVM. The results highlight the practicality of simple, interpretable machine learning models for efficient and reliable early-stage material selection and decision-making.</p>