Published Paper Details:

The proposed project aims to solve the fundamental problem of stress detection in IT professionals using imaging techniques. A recording of stressed and non-stressed people. Most importantly, the preprocessing stage includes transformation and gray transformation, while the extraction process includes both measurement methods such as index and standard deviation, and techniques such as local binary pattern (LBP). Then, various classification techniques were used, including neural network (CNN), support vector machine (SVM), and tree pruning models. The aim of this paper is to use machine learning and task visualization to identify overwork of IT personnel. Our technology is an improvement over traditional stress testing methods and does not involve a scale or personal counseling. This study is a modification of the anxiety research process that does not involve supervision or personal counseling. Gather information about employees' stress levels through research to provide effective solutions. This article explores stress management and how to create a healthy, productive work environment to get the most out of your employees. The model is trained and analyzed with a combination of SVM and decision-making models and accuracy measurements that are competitive with the CNN model. Accurately identifying stress situations provides an effective way to manage stress in the workplace.