Advanced Machine Learning Techniques for Predicting Health Status in Smoke-Exposed Environments

Smoker health status is a vital focus in public health and medical research due to smoking’s impact on preventable diseases and premature death worldwide. Recognizing the diverse effects of smoking on health is crucial, spanning both immediate and long-term consequences across multiple physiological systems. This research underscores the pressing importance of comprehending smoking’s far-reaching impacts on health and emphasizes the necessity of effective interventions and cessation strategies to mitigate its adverse effects. This study aims to utilize diverse machine learning algorithms to predict the health status of smokers accurately. The primary objective is to equip healthcare professionals with a reliable diagnostic tool for early and precise identification of smoking-related health risks, ultimately saving lives. Leveraging a comprehensive dataset, various algorithms, such as Boost, LightGBM, CatBoost, Random Forest, AdaBoost, and others, are explored. Through rigorous experimentation, the Boost algorithm stands out, achieving an impressive 96.2% accuracy in predicting the health status of smokers.