Abstract

Stress management is a growing concern in today's high-pressure tech industry, particularly for those working in information technology. Long hours, strict deadlines, and high expectations are common in the information technology workplace, all of which may contribute to an already stressful situation. Professionals' health, happiness, and productivity are all negatively impacted by unmanaged stress. In order to help with proactive stress management, this project intends to use machine learning methods to forecast the stress levels of IT workers. Heart rate, skin conductivity, hours worked, number of emails sent, and meetings attended are some of the characteristics that we use to identify job stress. Together, these characteristics illuminate the many ways in which stress manifests in the workplace, both physiologically and otherwise. Here, machine learning provides a fresh perspective on a problem that is becoming more pressing. This model's goal is to help people and businesses by using data analytics to deliver them actionable insights. Both individuals and organizations may benefit from these forecasts. The former can use them for self-monitoring and early intervention, while the latter can use them to understand which positions or settings cause the most stress and allocate resources accordingly. Our first findings show that the selected variables are significantly correlated with stress levels, suggesting that machine learning might be useful for predicting stress in IT workers. When it comes to workplace mental health and wellness, this report is a major step in the right direction.