On reading "Multimodal support to group dynamics"

As a student who is interested in intelligent systems, I'm more interested in how the system interprets the dialog in the meeting and inference each person's role in the meeting and provide feedbacks. However, this part is not described in the article, so I have no idea whether the mapping between group dynamics and computer generated suggestions is appropriate. I'm not worried that inference by computer would over simplify group dynamics or human personality, social science is very complex, computer science is complex, too. I think that wether the generated suggestions will help to improve the quality of meeting depends on the mapping process and how people use the feedback to adjust their behavior.

From the interaction point of view, this article made me think of a game: The Sims. In this game, players interact with virtual characters through a computer interface, the emotions and relationships are simplified and represented in a value. Each time the sim performs an action on another sim, it will get a brief feedback in a sentence or a plus/minus sign as the result/summary of that interaction.

All the relationship and personality parameters are closely linked to form the social dynamics in the game. It is really a fun game to play with, I feel that this game makes me think a lot about interactions between humans, and it also brings me some new point of view from its over- simplified but humorous interaction and emotion representation. I think the main idea of this paper, to provide a simplified feedback for a set of interactions to let the people have a better idea of the whole picture in order to improve or adjust actions, is somewhat like the game.

However, how much is the real life like the virtual game model is a difficult question to answer, so the effects of the proposed model might still depends on it real life performance.