Predictive Modelling for Lung Cancer Detection using Machine Learning Techniques
B. Yamini, K. Sudha, M. Nalini, G. Kavitha, R. Sıva Subramanıan, R. Sugumar
Abstract
Lung cancer (LC), generally called lung carcinoma, is a fatal and debilitating illness that affects the lungs. Lung cancer is caused by the genetic damage that occurs to the DNA of the cell as a result of numerous environmental factors. There are no common signs that can be used to determine if a man or a woman has lung cancer unless the illness has progressed to an advanced stage or a lung screening method has been utilized. Early detection and intervention will help save lives from the lung cancer epidemic. Based on this research problem, lung cancer prediction is investigated and tested. Several ML approaches are considered to the study of lung cancer. ML is an area of AI that seeks to generate predictions based on historical and past data. The experimental technique utilizes lung cancer data collected from the public respiratory system. Lung cancer datasets are analyzed using LR, DT, RF, SVM, XGB classifier, Gradient Boosting, and KNN models. Four distinct measures (accuracy, sensitivity, precision, and F1-score) are used to project and compare the experimental outcomes acquired by ML models: In relation Compared to other ML models such as LR, DT, RF, SVM, Gradient Boosting, & KNN, the experimental findings demonstrate that the XGB classifier provides accurate predictions. Finally from all parameter aspects, it clearly shows XGB classifier good prediction results compare to other ML models like LR, DT, RF, SVM, Gradient Boosting, and KNN.