EARLY DIAGNOSIS AND CLASSIFICATION OF LUNG CANCER DRIVEN BY MULTI-FEATURE DATA: A COMPARISON AND OPTIMIZATION OF THREE MACHINE LEARNING METHODS

The objective of this study was to construct a prediction model for early diagnosis and classification of lung cancer based on multi-dimensional clinical data. Three advanced machine learning models - Random Forest, Decision Tree, and Support Vector Machine (SVM) - were employed to predict lung cancer using features such as age, chronic disease history, and clinical symptoms. The models were optimized through various strategies to improve predictive performance. The results demonstrated that all three models achieved high accuracy and sensitivity in lung cancer prediction. Furthermore, a detailed analysis of feature importance identified key factors such as age and chronic disease history that significantly influenced prediction outcomes. Among the models, the SVM exhibited particularly strong performance, providing robust support for accurate lung cancer prediction. Future work will focus on integrating multimodal data and optimizing model architecture and hyperparameters to further enhance the predictive accuracy and clinical utility of the model, thereby contributing to the early diagnosis and treatment of lung cancer.