Lung Cancer Prediction based on Machine Learning

Abstract

Lung cancer, a malignancy with high incidence and mortality rates, underscores the critical importance of early diagnosis and treatment. Traditional prediction methods possess limitations, whereas advancements in machine learning technologies offer novel avenues for lung cancer prediction. This paper utilizes data from public databases of lung cancer patients, employing various machine learning algorithms such as Support Vector Machine (SVM), Random Forest (RF), Gradient Boosting Decision Tree (GBDT), and Neural Networks (NN) to diagnose and predict lung cancer. The results indicate that the Random Forest model performs optimally, particularly achieving a high AUC value. Notably, smoking status, age, and yellowing of fingers are identified as crucial features for lung cancer prediction. This study provides new insights and methods for early diagnosis and treatment of lung cancer, possessing significant clinical implications and application value.