Published Paper Details:

Stress disorders are a rather common occurrence among business sector workers. As people's job and lifestyles change, we can observe an increase in stress among working employees. Even while many business sectors offer a range of mental health programs and attempt to lessen stress-related problems in the workplace, the disorder is still very much alive and well. In this paper, we will apply two machine learning techniques to ascertain the level of stress experienced by corporate sector employees and attempt to identify the problems that contribute to the stress levels. Following the completion of data preprocessing and cleaning, we will use two machine learning approaches (SVM and Random Forest). Our trained model's accuracy was examined and evaluated in detail. Using these two machine learning techniques, it is discovered that the primary factors that lead to stress problems are an employee's sex, family background, and ease of access to health benefits at work. Because of these outcomes, corporate sectors can now reduce stress and create a very welcoming workplace for their personnel. In order to determine if an employee is in a relaxed or stressed state, the author uses social media datasets, such as tweets, where employees can express their opinions. However, manually assessing these opinions may require a significant amount of human labor. Therefore, we are utilizing machine learning techniques, and the results of our experiment indicate that these algorithms have an accuracy of over 80% in detecting stress.