Research and Application of Lung Cancer Diagnosis on Machine Learning Algorithm
Lung cancer is evil tumor which seriously harms human health and is famous for its high morbidity and mortality.Rapid and accurate diagnosis of lung cancer is a major challenge in the prevention and treatment of lung cancer,which is signifi-cance to human life and health.Support vector machine(SVM),random forest(RF)and XGBoost model are compared and analyzed.The accuracy,recall,f1 score,precision and ROC curve of the model are anlayzed,it is proved that the linear support vector ma-chine can better predict lung cancer,and the accuracy rate can reach 95.18%.The performance evaluation indexes of random forest and XGBoost model are highly improved on the data set balanced by SMOTE algorithm,and its accuracy can reach 89.16%and 90.36%.Random forest and XGBoost can get the prediction results faster than support vector machine,and they are also good model choices in the auxiliary diagnosis of lung cancer.
国家科技期刊平台
NSTL
万方数据
