Stress Detection of Office Employees Using Sentiment Analysis

Abstract

Due to the increasing competition in the industry, companies demand more work hours from employees, and employees take a lot of stress in completing their deadlines. Now with the existing deadline stress, they also face problems like family problems, low motivation, discrimination, politics, etc., which bring the extra negative stress that harms the productivity and mental peace of employees. To reduce workplace stress among the employees and increase productivity, there is a need for a system to identify the stress level so that remedial action can be taken beforehand. In this paper, we have proposed a method to detect the seven emotions (angry, disgust, happy, sad, fear, surprise, neutral) of employees at the workplace using facial expressions from the Web camera of their computers and sentiment analysis on the monthly reviews provided by the employees using natural language processing to calculate the stress level, and stress level is also calculated using the answer provided by the employee to the question “How was your day?” at the end of each day and generate a report for the human resources (HR) of the company who will analyze the stress level of the employees. HR can talk to them, counsel them, and help them which will ultimately motivate employees to do quality work.