Emotion recognition using musical mnemotherapy with three physiological sensors

Description

It exists some non-drug therapy relaying emotions to take care Alzheimer patients. This paper presents a new application for emotions detection. Indeed, a subject listen music to stimulate his emotional state and 3 physiological sensors are used to acquire his bio-signals for emotions detection. After constructing an emotion data base during the learning phase, we apply the recognition algorithm on each modality separately. Then, these decisions are merged separately by applying a decision level fusion. The experiments show the strength of our methodology to detect efficaciously these 3 emotional states when a subject is listening to music, indeed, we get a recognition rate of 84.245%

Abstract

International audience

Additional details