Studying the effectiveness of a model for predicting post COVID-19 stress among university students upon returning to campus

Abstract

Considering the COVID-19 pandemic, students have been engaged in remote learning for the past two years. This has resulted in changes in the way they learn and take exams. They have had to adapt to learning online, which has led to a lack of social interaction with peers and professors, causing elevated stress levels. Based on a survey by the Department of Mental Health, data analysis was conducted to determine the stress levels of students enrolled at Kasetsart University, Sriracha Campus using data mining techniques. Three models were used, including Decision Tree, Naïve Bayes, and K-Nearest Neighbor Algorithm. The results showed that Naïve Bayes had the highest accuracy in predicting stress levels among students after returning to university with a model efficiency of 80%. The findings of this research underline the importance for universities to provide support and resources for students who may be experiencing stress and anxiety related to the COVID-19 pandemic and remote learning. These services could include mental health services, academic guidance, and adaptable learning choices. Additionally, further research could be conducted to explore the effectiveness of different interventions and strategies for reducing stress levels and promoting well-being among university students during times of crisis and upheaval. This could include mindfulness-based practices, cognitive-behavioral therapy, and peer support initiatives.