REGRESSION-BASED PREDICTION OF ANXIETY SEVERITY

The present study explores the use of machine learning to predict self-reported anxiety levels based on demographic, behavioral, and physiological data. To this end, we used a dataset comprising 11,000 survey responses and applied multiple regression models, including Linear Regression, Gradient Boosting Regression, Extreme Gradient Boosting Regression Light Gradient-Boosting Machine Regression after data preprocessing. The performance of the models was evaluated using the mean absolute error, the root mean square error, and the coefficient of determination. Among the models evaluated, Gradient Boosting Regression achieved the best results, with a mean crossvalidated R² of 0.759 after five-fold cross-validation.