Comparison Of The Perfomance Of The C4.5 Anda Naïve Bayes Algorithms In Classifiying Lung Cancer

Abstrak

Lung cancer is one of the most dangerous diseases with a high mortality rate. Early detection is crucial to increasing the chances of recovery. This study aims to evaluate the performance of the C4.5 and Naïve Bayes algorithms in classifying lung cancer cases. The dataset used is a lung disease prediction dataset from Kaggle, consisting of 30,000 records and 9 attributes. The experimental process was carried out using RapidMiner with the 10-fold cross-validation method, and evaluation was performed using a confusion matrix and t-test. The results showed that the C4.5 algorithm achieved an average accuracy of 94.44%, while Naïve Bayes achieved 87.05%. The t-test result yielded a p-value of 0.000, indicating that the performance difference between the two algorithms is statistically significant, with C4.5 proving to be superior in classifying lung cancer cases. This research is expected to serve as a reference for the development of disease classification systems, particularly in assisting with early and more accurate lung cancer detection.