The paper "Towards Emotionally Intelligent Machines: Taking Social Contexts into Account" co-authored by IHPC's Dr Irene Lin and three NTU researchers received the Best Paper Award of the 18th International Conference on Human-Computer Interaction in the thematic area of "Social Computing and Social Media". This conference, well-regarded in the field of human-computer interaction, gives out one best paper award in each of the fifteen thematic areas within the conference. For the 2016 conference, best paper awards were given to 1.16% of the accepted papers.
The past decades saw increasing interest in incorporating emotional intelligence into machines as a way of improving human-computer interaction. While various computational emotion models had been developed, these models rarely account for the effect of social context on emotional understanding and expression.
This paper reports the team's findings that an individual's tendency for expressing positive emotions and negative emotions is affected by properties of the social network the individual belongs to. Such properties include number of friends that the individual has and the inter-connectivity among these friends. Importantly, the direction and magnitude of these effects was found to also vary with individual personality traits such as extraversion. Computational emotion models enhanced with such insights have various applications, including in conversational agents.
Read more about the work at this link