Creating and sharing knowledge for telecommunications

Applicability of Multi-Modal Electrophysiological Data Acquisition and Processing to Emotion Recognition

Canento, F. A. L. ; Silva, H. ; Fred, A. L. N.

Applicability of Multi-Modal Electrophysiological Data Acquisition and Processing to Emotion Recognition, Proc International Workshop on Computing Paradigms for Mental Health - MindCare, Vilamoura, Portugal, Vol. , pp. 59 - 70, February, 2012.

Digital Object Identifier:

Abstract
We present an overview and study on the applicability of multimodal electrophysiological data acquisition and processing to emotion recognition. We build on previous work in the field and further explore the emotion elicita- tion process, by using videos to stimulate emotions in several participants. Electrophysiological data from Electrocardiography (ECG), Blood Volume Pulse (BVP), Electrodermal Activity (EDA), Respiration (RESP), Electromyography (EMG), and Peripheral Temperature (SKT) sensors was acquired and used to classify the negative and positive emotions. We evaluate the emotional status identification accuracy both in terms of the target emotions and those reported by the participants, with recognition rates above 70% through Leave One Out Cross Validation (LOOCV) with a k-NN Classifier.