Uncovering Cervical Cancer High-Risk Factors Using HistGradientBoosting and SHAP-Driven Insights

Cervical cancer remains a significant global health concern, necessitating accurate and efficient predictive models for early risk identification. This study proposes an enhanced approach to cervical cancer risk classification by leveraging the HistGradientBoostingClassifier (HGB) in combination with advanced data imputation techniques and hyperparameter tuning. The methodology not only addresses missing data effectively but also optimizes the classifier’s performance through rigorous parameter optimization. To gain deeper insights into feature importance, the SHapley Additive exPlanations (SHAP) framework is employed, enabling the identification of key risk factors influencing predictions. The proposed model outperforms baseline classifiers, demonstrating superior accuracy, precision, and recall across all performance metrics. These results highlight the potential of our approach in improving predictive accuracy and uncovering critical cervical cancer risk factors, paving the way for more informed decision-making in clinical practice.