Comparative Analysis of Performance Between KNN and C5.0 Algorithms in Lung Cancer Disease Detection

Abstract

Lung cancer is a disease characterized by the growth of abnormal cells in the lungs that can spread to other parts of the body. In practice, medical teams will usually evaluate a patient's symptoms conventionally, which is highly inefficient and time-consuming, especially if there are a large number of patients. This manual evaluation process can cause delays in diagnosis and treatment, and increase the risk of errors. Therefore, this research will discuss lung cancer detection using the K-Nearest Neighbors (KNN) algorithm and the C5.0 algorithm in order to solve the problems previously described. The use of the K-Nearest Neighbors (KNN) algorithm and the C5.0 algorithm was chosen because these two algorithms have the ability to process complex data and produce accurate models. The results of this study will show a comparison of which performance is much better used to accurately detect lung cancer based on the amount of training data available, and it can be known that the lung cancer detection process can be done more quickly and efficiently, using the K-Nearest Neighbors (KNN) or C5.0 algorithm to improve diagnosis accuracy. The results show that the KNN algorithm is superior to the C5.0 algorithm specifically for lung cancer detection.