A short review on emotional recognition based on biosignal pattern analysis

Emotional intelligence has been argued to be more important than verbal or mathematical intelligence. So the last decade researchers try to embody the emotional with the artificial intelligence developing machines capable of understanding the human's emotions. This is the core of a modern and rapidly growing research field called Affective Computing. The main goal of the Affective Computing science is to understand the basis of emotions as well as the way they are expressed, and make the machines able to recognize the human's emotions. Recognizing emotional information requires the extraction of meaningful patterns from the gathered biological data. This is done by adopting machine learning techniques using one modality, or the fusion of different modalities. This study comes to introduce the physiological basis of emotional recognition in order to give us further evidence about the use of each modality, while it makes more clear the ways that someone can combine the data from different modalities. Moreover, the most prominent studies for emotional recognition, based only on biological signals, are reported, and despite the fact that their results are encouraging, there are some serious unsolved problems which are addressed and further discussed herein.