ENHANCING LUNGS CANCER PREDICTION USING ENSEMBLE TECHNIQUES WITH SUPPORT VECTOR MACHINES

Abstract

Lung cancer is still one of the leading types of cancer, the diagnosis of which in the early stages increases the likelihood of recovery. This paper thus aims at improving the prediction of lung cancer by using ensemble learning methods with Support Vector Machines (SVM). To enhance the SVM performance in predicting lung cancer from medical datasets, the research explores different ensembles methods such as boosting, bagging and stacking. The goals pursued in the assessments were to determine the accuracy, precision, recall, F1 score, and sensitivity of each of the models. The findings established that compared with baselines SVM, integrated with ensemble methods, particularly bagging and stacking offered enhanced prediction precision with sensitivity . The ensemble methods proved do enhance the ability of the model to perform well for imbalanced data and reduce overfitting. This study brings about the essence of identification of appropriate ensembles for the improvement of prediction consequence in clinical trials and the efficiency of ensemble SVM models.