Random Forest based Risk Factor Analysis for Lung Cancer Prediction

Early detection of lung cancer through risk factor analysis can significantly improve outcomes. This study presents a novel machine learning approach to analyze risk factors and predict lung cancer incidence using the Random Forest algorithm. A dataset of 1000 patient records with 21 attributes related to lifestyle, environmental exposure, and health conditions was utilized. Exploratory analysis revealed correlations between risk factors like smoking, air pollution, occupational hazards, and lung cancer, particularly in older males. The Random Forest classifier achieved remarkable performance with 99% accuracy, 100% recall, and 94% precision through k-fold cross-validation, outperforming traditional methods like Naive Bayes. Crucial risk factors were identified through feature importance analysis, with smoking, air pollution, occupational hazards, and genetic factors emerging as top predictors. The robust Random Forest model demonstrates efficacy in lung cancer risk prediction from diverse factors, facilitating early diagnosis and targeted screening strategies. This research highlights the potential of machine learning techniques, specifically ensemble methods like Random Forest, for robust risk factor analysis and accurate lung cancer prediction models to assist clinical decision-making.