A Systematic Method for Lung Cancer Classification
Ravi Kumar Sachdeva, Tushar Garg, Gagandeep Singh Khaira, Dikshant Mitrav, Rakesh Ahuja
Abstract
One of the top causes of mortality worldwide is lung cancer. The diagnosis procedure is unsatisfactory since lung cancer can only be found by a doctor when it is well advanced. To improve their accuracy and effectiveness, clinicians must apply machine learning algorithms that enable early lung cancer detection. The effectiveness of eight different classifiers, i.e., Naive Bayes (NB), Support Vector Machine (SVM), Decision Tree (DT), K Nearest Neighbors (KNN), Logistic Regression (LR), Random Forest (RF), Adaboost and Xgboost on the lung cancer dataset available on Kaggle, has been compared by authors to identify a systematic strategy for lung cancer classification. Accuracy, Sensitivity, Specificity, Precision, and F-Measure have been used as performance parameters. Among all applied classifiers, NB had the highest accuracy (98.33%).