Applying Machine Learning Techniques and Feature Selection Methods for Lung Cancer Prediction

This research aims to apply several machine learning techniques, including Neural Networks, Random Forests, Naive Bayes, K-nearest Neighbors, and Decision Trees, to predict lung cancer. A dataset from the Kaggle database, consisting of 309 records and 16 features, was used to evaluate the performance of these ML algorithms. The models were compared based on their effectiveness with different feature selection methods. The results of the research showed that the Random Forest model was the most effective for predicting lung cancer. The Information Gain approach identified the 14 most suitable features and achieved the highest accuracy of 96.77%, precision of 96.80%, recall of 96.80%, f-measure of 96.80%, and RMSE of 0.2113. These results demonstrate how the proposed approach can improve prognostic and diagnostic systems for lung cancer treatment planning and decision-making. Early diagnosis and treatment of lung cancer can significantly reduce mortality rates.