<p>Dual-phase steels are among the essential class of engineering materials in metallurgical industries that find enormous applications. Predicting mechanical properties, including yield strength (YS), ultimate tensile strength (UTS), total elongation (TE) and uniform elongation (UE), based on processing parameters and microstructural variables is crucial for enhancing their reliability. This study utilizes a comprehensive dataset of 82 data points sourced from the existing literature predict the mechanical properties of dual-phase steels using machine learning techniques, even with a relatively small dataset. There were three groups into which the input data were divided: Class I: composition (%Nb, %C, %Mn, %Si, %V, %Ti), Class II: microstructural features (martensite fraction, ferrite fraction, ferrite grain size), and Class III: processing parameters (intercritical annealing time, temperature, and amount of deformation). Five machine learning models, comprising Support Vector Regressor, Decision Trees, Random Forest, Adaptive Gradient Boost and Extreme Gradient Boost, were trained and optimized for robust predictions. The xGBoost algorithm achieved the best prediction accuracy for YS and UTS, with R² values of 0.991 and 0.976, respectively, while Random Forest provided superior predictions for TE and UE with R² values of 0.970 and 0.960, respectively. These results demonstrate excellent generalization and learning capabilities, exhibiting minimum Mean Squared Errors and execution times below 0.7 seconds for all predictions. The findings highlight the potential of ML-based approaches to optimize processing parameters for DP steels, enabling the precise tailoring of microstructures to achieve desired mechanical properties. This study establishes a pathway for further advancements in integrating material science and artificial intelligence, driving innovations in steel manufacturing and Industry 4.0.</p>