This study proposes a computational framework for providing affective labels to real-life situations. We first define an affective situation as a specific arrangement of affective entities relevant to emotion elicitation in an affective situation. Then, an affective situation is represented as a set of labels in the valence-arousal emotion space. Based on physiological behaviors in response to a situation, the proposed framework quantifies the expected emotion evoked by the interaction with a stimulus event. The accumulated result in a spatial-temporal situation is represented as a polynomial curve called affective curve. It bridges the semantic gap between cognitive and affective perception in real-world situations. We show the efficacy of the curve on reliable emotion labels in real-world experiments, concerning 1) a comparative study between the results from our system and existing explicit assessment on
measuring emotion, 2) physiological distinctiveness in emotional states, and 3) physiological characteristics correlated to continuous labels.
[AI 사피엔스 시대]”에이, 짜증나” “아이, 행복해”…인간과 ‘희노애락’ 나누다
Related publications
1. BH Kim, S Jo, S Choi, A-Situ: a computational framework for affective labeling from psychological behaviors in real-life situations, Scientific Reports, 10, 15916, Sept 2020 [LINK]
