DETECTION OF STRESS IN IT EMPLOYEES USING DEEP LEARNING

Abstract

In the contemporary era, characterized by advanced technological devices, a pervasive sense of stress is increasingly affecting individuals. Despite the abundance of material wealth, people often find themselves discontented. Stress, defined as a feeling of pressure, can manifest in mental, emotional, or physical forms. It is crucial to establish effective stress management systems to gauge and address stress levels that disrupt our socio-economic lifestyles. According to the World Health Organization (WHO), one out of every four individuals grapples with the mental health challenge of stress.The stress experienced by individuals can lead to both mental and socioeconomic challenges, including decreased concentration at work, strained relationships with colleagues, feelings of despair, and, in extreme cases, even suicide. To address this, it is essential to offer counseling support to individuals facing stress, aiding them in effectively managing their emotional burden. While it's unrealistic to eliminate stress entirely, adopting preventive measures can play a crucial role in its effective management.Only individuals with medical and physiological expertise are currently able to evaluate if someone is experiencing depression or stress. One well-established method for detecting stress involves using questionnaires. The main objective of our approach is to employ advanced deep learning and image processing techniques to recognize indicators of stress in IT professionals.This technology represents an enhanced iteration of prior stress detection technologies, distinguishing itself by incorporating considerations for employee emotions and real-time detection. In contrast to its predecessors, this system integrates both periodic and immediate detection of employee emotions.The identification of stress through automated means reduces the likelihood of health problems and enhances the well-being of both the IT employee and the organization. Understanding the emotional state of IT employees enables the company to offer appropriate support, leading to improved performance. The suggested system model, constructed with CNN Model Architecture demonstrates an accuracy of 98.45% and with In an accuracy of 70%