Smoking stands as a prominent preventable hazard that contributes to untimely mortality. Smoking remains a concern for health among students as it has enduring impacts on both personal well being and society, at large. For the purpose of creating focused treatments and preventive programs, it is essential to identify the variables that affect students' propensity to start smoking. This study uses machine learning regression approaches to give a thorough analysis of predicting student smoking propensity. We acquired a variety of data from a sizable number of students who were enrolled in educational institutions. Details about their demographics, socioeconomic situation, academic accomplishments, peer influences, and various facets of their lifestyle are included in this data. We utilized this dataset to make predictions on the probability of students engaging in smoking. In our research we employed regression models such, as Linear regression, Ridge regression, Lasso regression, Decision tree regression etc. Keywords—PCOS, Machine Learning, Classification, SVM, KNN, XGBoost, Logistic Regression, Random Forest.