*Result*: Application of machine learning algorithms for predicting mechanical properties of stainless steel.

*Materials informatics is a novel approach in material science that combines information technology and material science to enhance efficiency and innovation in the discovery of new materials. In materials informatics, experimental and simulation data are merged with data-driven methods like big data and machine learning to gain a deeper understanding of material properties. In this study, the performance of six machine learning algorithm models is compared for predicting the mechanical properties of austenitic stainless steel (ASS), such as Yield strength (YS), Ultimate tensile strength (UTS), and Elongation (EL), based on chemical composition and heat treatment temperature. Each model is evaluated using metrics such as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R-squared. The results of the model testing demonstrate that the Gradient Boosting model exhibits the best performance in predicting the mechanical properties of stainless steel. This model provides predictions with low levels of error, as indicated by the smaller MAE and RMSE values for all three mechanical properties. Furthermore, the Gradient Boosting model effectively explains data variation, as evidenced by the high R-squared value. The study also involves model validation using new data not present in the training dataset. The testing with this new data reaffirms that the Gradient Boosting model consistently delivers accurate predictions. This can be observed from the MAE value of 9.86, RMSE of 14.37, and R-squared of 0.86 for YS. For UTS, the model achieves an MAE of 13.46, RMSE of 16.47, and R-squared of 0.95, while for EL, it obtains an MAE of 3.13, RMSE of 4.47, and R-squared of 0.76. [ABSTRACT FROM AUTHOR]*