Emotion Recognition of Physical Activities for Health Monitoring Using Machine Learning

Emotion recognition based on physiological data has attracted increasing attention in physiological monitoring, affective computing, and other fields. This paper proposes a method to classify human's emotion for health monitoring in physical activities by using machine learning. Participants completed the experiment including walking, running, and other physical activities. The data of photoplethysmography (PPG) and electrodermal activity (EDA) were recorded by wearable sensors on participants. After the data processing and feature extraction, two classifiers, support vector machine (SVM) and random forest (RF) were applied independently on the dataset to classify human's emotion, including calm, excited, relaxed, bored, and afraid. As a result, the SVM classifier achieved an accuracy of 81.87% and the accuracy of RF classifier is 86.61%. These results demonstrated the effectiveness of the proposed method on emotion recognition in human's physical activities.