Lung Cancer Survivability Prediction based on Performance Using Classification Techniques of Support Vector Machines, C4.5 and Naive Bayes Algorithms for Healthcare Analytics

• Published in: Procedia computer science Vol. 132; pp. 412 - 420
• Main Authors: K R, Pradeep, N C, Naveen
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 2018
• Subjects: C4.5; HcA; Lung Cancer; NBs; SVM; Lung Cancer; SVM; NBs; HcA; C4.5; ML
• ISSN: 1877-0509; 1877-0509
• DOI: 10.1016/j.procs.2018.05.162

The Healthcare Analytics(HcA) is a process in which clinical data is analyzed and patient’s treatment is performed. The treatment depends on the analysis of clinical data accumulated from Electronic Health Records (EHRs), pharmaceutical and research and development cost and claims of patient. Lung cancer is the most common among cancer disease and the foremost reason for deaths in both men and women. In this research work EHRs are analyzed and the survivability rate is predicted for lung cancer. Researchers apply Machine Learning Techniques (MLT)for predicting the survivability rate so that chemotherapy can be provided for cancer affected people. MLTare well accepted by doctors and work well in diagnosing and predicting cancer. An ensemble of Support Vector Machine (SVM), Naive Bayes (NBs)and classification trees (C4.5) can be used to evaluate patterns that are risk factors for lung cancer study. The North Central Cancer Treatment Group (NCCTG) lung cancer data set along with new patient data is used for evaluating the performance of support SVM, NBs and C4.5. The comparison isbased on accuracy, Area Under the Curve(AUC), Receiver Operating Characteristic (ROC) and the resultshows that C4.5 performs better in predicting lung cancer with the increase in training data set.