Recognizing Emotion from Blood Volume Pulse and Skin Conductance Sensor Using Machine Learning Algorithms

Abstract

Recognizing emotional states is becoming a major part of a user’s context for wearable computing applications. The system should be able to acquire a user’s emotional states by using physiological sensors. We want to develop a personal emotional states recognition system that is practical, reliable, and can be used for health-care related applications. We propose to use the eHealth platform [1] which is a ready-made, light weight, small and easy to use device for recognizing a few emotional states like ‘Sad’, ‘Dislike’, ‘Joy’, ‘Stress’, ‘Normal’, ‘No-Idea’, ‘Positive’ and ‘Negative’ using decision tree (J48) and IBK classifiers. In this chapter, we present an approach to build a system that exhibits this property and provides evidence based on data for 8 different emotional states collected from 24 different subjects. Our results indicate that the system has an accuracy rate of approximately 92%. In our work, we used two physiological sensors (i.e. ‘Blood Volume Pulse’ and ‘Galvanic Skin Response’) in order to recognize emotional states (i.e. stress, joy/happy, sad, normal/neutral, dislike, no-idea, positive and negative).