A Comprehensive Framework on Emotion-Aware Stress Detection Using IoT and Machine Learning

Abstract

In today's society, the chapter thrives on performance, competition, and perfection, which leads to an insidious increase in stress. Stress is a condition of mental pressure for particular individuals facing problems from environmental and social well-being, which leads to so many diseases. Stress causes damage that is often underestimated, and it is a social phenomenon that should be closely examined and evaluated. Nowadays, stress leads to suicide attempts, depression, mental conditions, etc. Using IoT sensors like GSR MAX30105 and IoT components such as Node MCU, Arduino UNO, and LCD can detect stress. Stress can also be measured through parameters like Spo2, HR/PR, and temp levels. The internet of things (IoT) and machine learning (ML) alarm the situation when the person is at real risk. ML is used to predict the condition of the patient by using techniques like SVM (support vector machines), CNN (convolutional neural networks), k-nearest neighbours, and logistic regression, etc. The user will get feedback through the GSM module and the internet based on the above information. According to the latest findings in the literature, an IoT-based system is designed to monitor the well-being of people during their daily activities. By capturing human facial expressions through the camera, emotion is detected. Finally, by combining stress and emotion detection through an efficient ML algorithm, the stress levels as low, medium, or high stress are predicted.