The research focuses on predicting interpersonal affect in human-robot interactions. The study involved 30 participant dyads who rated their perception of the other interactant after a short interaction using the CORAE tool. The study found a significant correlation between these interpersonal ratings and aspects such as personality traits, participation balance, and sentiment analysis. One notable finding was the significant effect of conversational imbalance on the retrospective ratings. The research contributes to enhanced human-robot interactions by understanding affect as a dynamic and context-dependent outcome.
Publication date: 17 Nov 2023
Project Page: 32nd IEEE International Conference on Robot and Human Interactive Communication
Paper: https://arxiv.org/pdf/2311.09378
