<p>This study investigates how doping various elements from the periodic table affects the mechanical properties of Fe-Cr-Ni-Al high-entropy alloys using machine learning techniques. Dimensionality reduction was applied to identify 26 representative elements, which were subsequently used to establish doping structure models. The elastic constants and intrinsic mechanical properties of these alloy configurations were evaluated using first-principles calculations. Chemical compositions were converted into physical features, serving as input variables for multiple machine learning algorithms to predict the properties of alloys with common dopants. Furthermore, elements were clustered according to their influence on alloy properties. The results reveal significant variability in the effects of different elements. Notably, Young’s modulus and toughness often exhibited opposing trends. For instance, Zn and Co enhanced toughness, whereas Li and Pb led to increased brittleness. Meanwhile, elements such as Pt, Ti, and Mn achieved a favorable balance between stiffness and ductility. Comparisons between experimental data and predicted results confirmed the accuracy of our approach. This method provides theoretical guidance for the compositional design of high-entropy alloys and offers insights into accelerating first-principles calculations.</p>