Predictions of College Students’ Mental Stress using Machine Learning Algorithms

Mental stress is becoming a big problem, particularly among younger individuals. The generation, which was previously regarded as being the easiest going is currently experiencing a great deal of stress. Stress can cause a number of problems every day, notably anxiety, suicide, cardiac arrest, as well as strokes. This study measures the students’ mental stress a week prior to their examination and while participants utilize the internet. Analysis of stress among college students at various stages of life is the proposed research goal, the stress may be due to exam stress or recruitment stress on the student. This study has analyzed how these influence a student’s cognition and examine the relationship between stress and online time. Machine learning technology is crucial for categorizing and predicting anxiety levels during various stages. Many machine learning approaches, such as a Decision tree, Support vector machine, Nave Bayes, and KNN have been employed to predict college students’ levels of stress. Input for this research was gathered from students at several institutions through surveys sent to dissimilar participants. The predicted outcomes have been assessed using many metrics, including accuracy, precision, and recall. To conclude, the KNN achieves the highest accuracy (87.2%).