Unmasking of Agony in Software Engineers Applying Machine Learning Approach
Y. Chitty
PG Scholar, Department of Computer Science & Engineering, Guru Nank Institutions Technical Campus Hyderabad.
Dr. Geeta Tripathi
Professor & HOD Department of CSE, Guru Nanak Institutions Technical Campus, Hyderabad.
Abstract: In the modern world with latest technology gadgets, stress is raising most to everyone. Because of this, despite affluence, people are not satisfied. A pressured feeling is stress. Pressure may be mental, emotional, or even physical. Systems for managing stress are essential for identifying the stress levels that disturb our socioeconomic way of life. According to the World Health Organization (WHO), one in four people suffer from the mental health issue of stress. Human stress causes mental and socioeconomic issues, loss of focus at work, strained relationships with coworkers, despair, and in the worst circumstances, suicide. This requires the provision of counseling to help those under stress manage their stress. While it is impossible to completely avoid stress, taking preventive measures can help you manage it. Only medical and physiological professionals can now assess whether a person is depressed or stressed. A questionnaire-based approach is one of the more established ways to identify stress. Our project’s primary goal is to identify signs of stress in IT professionals utilizing sophisticated machine learning and image processing methods. Our technology is an improved version of the previous stress detection technologies, which did not take into account the employee’s emotions or live detection. However, the system includes both periodic and live employee emotion detection. Automatic detection of stress minimizes the risk of health issues and improves the welfare of the IT employee and the company. Knowing the IT employee’s emotions allows the business to provide the right guidance and obtain better results from them. The accuracy of our suggested system model, which is developed using CNN Model Architecture, is 87.34% during training and 98.45% during validation.
Keywords: CNN, Stress Detection, Machine Learning.
