Rushikesh Muley

Lead predictive analysis efforts on steel fatigue strength, achieving 94.44% accuracy using PCA and Ridge Regression. Conducted extensive statistical analysis including Cook’s Distance, Durbin-Watson, and VIF as part of ongoing DRDO research. Developed a property prediction model for hot rolled steel employing Boruta Shap for feature selection, Random Forest and XGBOOST algorithms for predicting mechanical properties. This work was published in the Machine Learning Research Journal.

Conducted predictive analysis of steel fatigue strength achieving 94.44% accuracy using PCA and Ridge Regression, with statistical validation through Cook’s Distance, Durbin-Watson, and VIF. Developed a property prediction model for hot-rolled steel using Boruta Shap for feature selection; applied Random Forest for tensile and yield strength prediction, and XGBOOST for elongation. This research was accepted by the Machine Learning Research Journal.

Led predictive analysis projects on steel fatigue strength achieving 94.44% accuracy using PCA and Ridge Regression. Conducted thorough statistical analyses employing Cook’s Distance, Durbin-Watson, and VIF techniques. Developed property prediction models for hot rolled steel using Boruta Shap for feature selection and machine learning algorithms including Random Forest and XGBOOST for mechanical property predictions. Published research in a machine learning journal and contributed to ongoing DRDO research initiatives.