Classification Model for the Detection of Anxiety in University Students: A Case Study at UNMSM

Anxiety, a common mental disorder worldwide, affects millions of people, including university students. In Peru, in 2021, a high prevalence of anxiety of 69% was recorded among students at three institutions of the Consortium of Universities. Early detection and treatment of anxiety in academic settings are critical to ensure student well-being. This study proposes a classification model for which Machine Learning algorithms such as Random Forest (RF), Naive Bayes, and Support Vector Machine (SVM) were evaluated using the Universidad Nacional Mayor de San Marcos as a case study. Advanced techniques were implemented, including Adversarial Generative Networks data augmentation, and hyperparameter tuning was performed using GridSearch and Random Search. The results highlighted the SVM model, which achieved 85% accuracy and an F1-Score of 0.78, with a prediction time of 0.00100159 s. Notably, the data augmentation technique revealed a steady improvement in all models.