Decoding Sleep Patterns through Explainable AI An Interaction Aware Hive Ensemble Model

Sleep quality, affected by lifestyle, stress, and physiological aspects, is an important indicator of overall well-being. In order to forecast sleep health, this research proposes the, Explainable Interaction Aware Hive Ensemble Framework (EIA-X). The model incorporates SMOTE balance, polynomial feature interactions, and several base learners such as Random Forest, Gradient Boosting, LightGBM, Logistic Regression, and XGBoost—into a hive ensemble method. Based on experimental findings, baseline models such as LightGBM (0.907), Gradient Boosting (0.893), and Random Forest (0.89) are outperformed by EIA-X in accuracy (0.920). Moreover, EIA-X improves progressively with respect to ROC-AUC and F1-score metrics, indicating generalizability and strength. Integration of Explainable AI offers interpretative insights into feature influence, making the proposed model a clear, effective framework for decision support and individualized sleep health prediction.