Toward Machine Emotional Intelligence: Analysis of Affective Physiological State
IEEE Transactions on Pattern Analysis and Machine Intelligence - Graph Algorithms and Computer Vision
Using noninvasive wearable computers to recognize human emotions from physiological signals
EURASIP Journal on Applied Signal Processing
International Journal of Human-Computer Studies
Hi-index | 0.01 |
The Human++ program aims at achieving highly miniaturized, wireless, intelligent and autonomous body sensor nodes to assist our health, comfort and lifestyle. In this paper the concept of body area network is applied to wireless monitoring of emotions, thus opening a new, affective, dimension in human computer interaction. A prototype body area network targeting the monitoring of physiological responses from the autonomous system is introduced, and tested for the classification of discrete emotions. Using data fusion and regression analysis, we show that the wireless physiological data can be mapped in real-time to an estimation of an individual's arousal level. Results in a controlled environment are presented, and specific challenges that need to be overcome for a widespread use of the technology are discussed. Finally, we show how advances in micro-power generation devices may lead to fully autonomous systems in the future.