Contenu principal de l'article
Mind Stress is a critical issue in today’s society, the impacts of which are observed mainly among youngsters aged 15 to 28. At this vulnerable age, most youngsters are college students and are going through a defining phase of their lives. The students who are perceived as careless may in fact be going through a stressful phase. There are many factors that can induce stress in a student’s life. Some of them are the pressure of exams, not getting a job, or due to relationship issues with friends. Students may get addicted to alcohol, drugs, and may even attempt suicide to find refuge from their stress. In this paper, an objective is set to analyze the stress levels in college students using data analytics and various machine learning models, and to predict students with high risk factors. A dataset containing students' responses for the Perceived Stress Scale test is collected, and on this dataset, the chosen machine learning algorithms are applied. A K-fold cross-validation optimization is applied to the chosen algorithms. Comparison for these models with and without the K-fold cross-validation optimization on the dataset are presented and discussed.