Comparison of Machine Learning Model for Lung Cancer Detection Using Classification Techniques

Lung cancer is one of the most common types of cancer that causes death in the world. In 2020, lung cancer had caused 1,796,144 deaths. This research provides a solution for lung cancer detection using machine-learning classification methods. This research aims to determine the performance of classification algorithms used for building classification models for lung cancer detection. Those algorithms include Support Vector Machine, Decision Tree, Random Forest, and Naïve Bayes. The Support Vector Machine algorithm provides performance with an accuracy score of 96.5%, a precision score of 95.83%, and a recall score of 96.66%. The Decision Tree algorithm came up with a performance with an accuracy score of 92%, a precision score of 92.18%, and a recall score of 94.9%. The Random Forest algorithm performs with an accuracy score of 95.5%, a precision score of 95.62%, and a recall score of 94.9%. Moreover, the Naïve Bayes algorithm provides performance with an accuracy score of 80%, a precision score of 79.69%, and a recall score of 80.17%. The comparison of the performance of those four algorithms shows that the Support Vector Machine algorithm provides the best performance in building classification models for lung cancer detection.