Predicting Lung Cancer using XGBoost and other Ensemble Learning Models
Divanshu Singh, Avish Khandelwal, Pawan Bhandari, Sunita Barve, Diptee Chikmurge
Abstract
Abnormal reproduction of cancerous cell growths in the lungs leads to highly fatal lung cancer. Early detection is essential for reducing death rates of lung cancer. New research has shown that machine learning algorithms can effectively diagnose lung cancer from medical datasets. This paper reviews the predictive effectiveness of various machine learning ensemble techniques used in lung cancer detection, like SVM, XGBoost, LightGBM, AdaBoost, CatBoost, and Random Forest with their performance analysis. Boosting techniques AdaBoost and XGBoost outperformed other algorithms with accuracy scores of 96.77% and 96.76% respectively. Therefore, we conclude that machine learning-based techniques hold great promise for improving lung cancer diagnosis and reducing mortality rates.