Risk Factors to Diagnosis A Data-Driven Model for Lung Cancer Prediction using BES-DT

2026-08-11T21:34:13+07:00

Lung cancer remains one of the most critical health challenges worldwide, requiring accurate and interpretable predictive models to support early diagnosis. This study proposes a novel framework by integrating the Bald Eagle Search (BES) optimization algorithm with the Decision Tree (DT) classifier, forming the BES-DT model. The dataset, consisting of demographic, behavioral, and clinical risk factors, was preprocessed and optimized using BES to select the most informative features while reducing redundancy. BES, inspired by the hunting behavior of bald eagles, balances exploration and exploitation to identify optimal solutions, enabling DT to construct transparent decision rules for clinical interpretation. Experimental evaluation demonstrated that BES-DT achieved superior performance compared to the baseline DT, with accuracy of 92.6%, precision of 98.8%, recall of 93.0%, F1-score of 95.8%, and ROC-AUC of 0.90, confirming strong discriminative ability. These findings highlight that BES-DT not only improves predictive accuracy but also maintains interpretability, offering a balanced solution for medical decision support. In conclusion, BES-DT shows significant potential in advancing early detection strategies for lung cancer.

Journal of Software Engineering Ampera

Risk Factors to Diagnosis A Data-Driven Model for Lung Cancer Prediction using BES-DT