Personal tools
You are here: Home Members Reza Amini Research


Posted by Reza Amini |

Reza Amini
Reza Amini
Key research interests:

Affective Computing; Modeling Empathy and Rapport for Virtual Agents; Humanization of Virtual Characters.

My research focus is on the design of intelligent virtual agents who can be expressive, culturally sensitive, and socially appropriate, in other words, humanized virtual agents.

Humanized socially intelligent agents sense their interlocutor’s social cues, including emotionmood, and personality, and respond to them correspondingly.

Recently I’ve been designing and evaluating my agents specifically in domains such as personal health informatics, social skills training, health education, health promotion, and automation systems. However, there are many other application domains where socially expressive virtual agents are of interest, e.g. virtual learning environments, car safety, social robotics, and tele-home healthcare.

Affective Computing is an inter-disciplinary research area, in which we study and synthesize social knowledge and turn them into computational models useful for affective intelligent virtual agents. We research different Artificial Intelligent (AI) techniques for various components of the intelligent agents (from sensing, to decision-making, to actuating), and apply Human-Computer Interaction (HCI) principles toward the design of engaging interactive and affective media.

In other words, I build affective intelligent virtual agents able to:

  • sense the affect (e.g., emotionmood), preferencesvalues, and personality of their interlocutor (through visual channels, text, and speech) and model the interlocutor’s features based on well-known computational model;
  • make decisions (using machine learning, data mining, logic, and probabilistic reasoning techniques) that are socially acceptable based on their dynamic user-model (knowledge representation);
  • interact empathetically with humans (HCI design principles) within the domain knowledge (e.g. health interventions, tutoring system);
  • build rapport, display emotion, and show social competence (based on emotion and social communication theories); and
  • learn to tailor and adapt (machine learning) their interactive styles to the specific socio-emotional profile (user-modeling) of their human counterpart.
Document Actions
Powered by Plone

Portal usage statistics