<![CDATA[Smoking behavior within households is a significant public health issue, particularly in densely populated regions such as West Java. This study aims to classify household smoking status using two modeling approaches, namely Logistic Regression and XGBoost, under several data-handling schemes including baseline models, One-Hot Encoding, and Weight of Evidence (WoE). Model performance is evaluated using accuracy, sensitivity, specificity, and Area Under the Curve (AUC) metrics. The evaluation results show that XGBoost with WoE Encoding provides the best performance, achieving an accuracy of 0.739 and sensitivity of 0.782, making it more effective in detecting smoking households compared to other methods. Feature importance analysis identifies the Food–Non-Food Expenditure Ratio (X1) as the most influential predictor. The dominance of this expenditure ratio confirms that household expenditure structure, as an indicator of economic condition and welfare, plays a crucial role in distinguishing smoking households. These findings support the hypothesis that poorer households tend to have a higher probability of smoking.]]>
Copyrights © 2026
