Identifying factors associated with vaping cessation in young adults: A machine learning and XAI approach
This study uses machine learning and explainable AI to identify factors predicting vaping cessation in young adults.Summary
This study investigates the factors associated with vaping cessation among young adults using machine learning (ML) and explainable artificial intelligence (XAI). The researchers conducted a social media-based survey to collect data on behavioral, contextual, and demographic factors from 119 participants. The study evaluated various predictive models, including linear models like LASSO, ridge regression, and elastic net, as well as non-linear models like random forest and support vector machines. The results indicate that linear models generally performed better and showed less overfitting than non-linear models. Key predictors identified for successful cessation included age, environmental triggers (particularly social exposure), vaping frequency, sex, and long-term behavioral outlook. The study highlights the importance of behavioral and contextual factors in cessation but notes that the findings are exploratory due to the cross-sectional design and sample size limitations. The authors recommend larger, longitudinal studies to validate these insights and inform targeted public health interventions.
(Source：PLOS (Public Library of Science))
