Analyzing Factors Influencing Temporary and Permanent Work-Related Accident Disabilities: An Interpretable Machine Learning

Project Code :TCMAPY2110

Objective

This study explores the application of machine learning techniques to predict work-related accident outcomes, specifically distinguishing between temporary and permanent disabilities, using a CSV dataset containing various features related to workplace injuries. Feature selection methods were applied to identify the ten most influential predictors, which were then used to train several classification models, including CatBoost, AdaBoost, XGBoost, and Extra Trees Classifier. Among these, the Random Forest (RF) model achieved the highest performance, recording an accuracy of 99.01% and an F1-score of 86.59%, significantly improving upon the baseline model with an accuracy of 66.62%. The results emphasize the potential of machine learning in accurately predicting accident outcomes, thereby identifying early risk factors and supporting targeted health interventions in occupational settings. The study also highlights the importance of model interpretability to ensure informed decision-making and the implementation of effective safety measures based on predictive insights.

Abstract

This study explores the application of machine learning techniques to predict work-related accident outcomes, specifically distinguishing between temporary and permanent disabilities, using a CSV dataset containing various features related to workplace injuries. Feature selection methods were applied to identify the ten most influential predictors, which were then used to train several classification models, including CatBoost, AdaBoost, XGBoost, and Extra Trees Classifier.recording an accuracy of 99.01% and an F1-score of 86.59%, significantly improving upon the baseline model with an accuracy of 66.62%. The results emphasize the potential of machine learning in accurately predicting accident outcomes, thereby identifying early risk factors and supporting targeted health interventions in occupational settings. The study also highlights the importance of model interpretability to ensure informed decision-making and the implementation of effective safety measures based on predictive insights.

Keywords: Accident prevention, machine learning, occupational health, occupational safety, predictive models.

NOTE: Without the concern of our team, please don't submit to the college. This Abstract varies based on student requirements.