Integrated Explainable Ensemble Machine Learning Prediction of Injury Severity in Agricultural Accidents

Agricultural injuries remain a significant occupational hazard, causing substantial human and economic losses worldwide. This study investigates the prediction of agricultural injury severity using both linear and ensemble machine learning (ML) models and applies explainable AI (XAI) techniques to understand the contribution of input features. Data from AgInjuryNews (2015-2024) was preprocessed to extract relevant attributes such as location, time, age, and safety measures. The dataset comprised 2,421 incidents categorized as fatal or non-fatal. Various ML models, including Naive Bayes (NB), Decision Tree (DT), Support Vector Machine (SVM), Random Forest (RF), and Gradient Boosting (GB), were trained and evaluated using standard performance metrics. Ensemble models demonstrated superior accuracy and recall compared to linear models, with XGBoost achieving a recall of 100% for fatal injuries. However, all models faced challenges in predicting non-fatal injuries due to class imbalance. SHAP analysis provided insights into feature importance, with age, gender, location, and time emerging as the most influential predictors across models. This research highlights the effectiveness of ensemble ML models in injury prediction while emphasizing the need for balanced datasets and XAI techniques for actionable insights. The findings have practical implications for enhancing agricultural safety and guiding policy interventions. HighlightsO_LIThis study analyzed 2,421 agricultural injury incidents from AgInjuryNews (2015- 2024) and utilized machine learning models to predict injury severity, focusing on both fatal and non-fatal outcomes. C_LIO_LIEnsemble models, such as XGBoost and Random Forest, outperformed linear models in accuracy and recall, especially in predicting fatal injuries, although challenges in non-fatal predictions due to class imbalance were observed. C_LIO_LIKey predictors identified through SHAP analysis included age, gender, location, and time, providing interpretable insights into the factors influencing injury severity. C_LIO_LIThe integration of explainable AI (XAI) enhanced the transparency of machine learning predictions, enabling stakeholders to prioritize targeted safety interventions effectively. C_LIO_LIThis research highlights the potential of combining ensemble ML models with XAI techniques to improve agricultural safety practices and provides a foundation for addressing data challenges in future studies. C_LI Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=142 SRC="FIGDIR/small/25321769v1_ufig1.gif" ALT="Figure 1"> View larger version (60K): org.highwire.dtl.DTLVardef@dc19a8org.highwire.dtl.DTLVardef@189646org.highwire.dtl.DTLVardef@31d6f2org.highwire.dtl.DTLVardef@16b01f_HPS_FORMAT_FIGEXP M_FIG C_FIG