Using Holonic Multi-agent Architecture to deal with complexity in Multi-modal emotion recognition

The affective computing is an increasingly important area in computer science. It aims to give computer system the ability to recognize and express affect. A key issue in affective computing is emotional data. Data in multiple formats and from different sources can be used and every format can be subject of different approaches and techniques. Several approaches can be combined to obtain better results. Because of this, multi-modal emotion detection requires an architecture that can support complexity and heterogeneity. This paper proposes a Holonic multi-agent architecture to design multi-modal emotion recognition and affective computing systems. Also, it describes the design and implementation of a prototype that demonstrates the applicability, adequacy and adaptability of the proposed architecture.