Smoking is considered as one of the most common causes of death and diseases around the world. Its negative effect on the health of individuals and groups leads to preventable mortality and morbidity in many countries. This study aims to utilize bio-signals and their related features to predict the smoking status among patients using machine learning. The main part of this study is to use two different methods of features selection: Principal component analysis (PCA) and Ant Lion Optimizer (ALO) followed by a comparison of the effectiveness between both two methods. we used: accuracy, sensitivity, specificity, and precision measures for comparison, and the results showed that ALO algorithm outperformed with an accuracy of 98.1%, a sensitivity of 98.2%, a specificity of 97.9%, and a precision of 97.9%.
