Modelling and statistical analysis of emotions in 3D space

Abstract

Emotional Intelligence provides an impetus for simulating human emotions in systems to make emotionally-sensitive machines. Integrating emotion-based theories and principles maturing with research in affective computing, we propose a novel statistical approach that can evaluate the correlation between different emotional states. It provides a way specialists can address the development of the entire passion experience, as reviewed through self-report. We also represent a three-dimensional model that can accommodate affect variabilities and analyze the distribution of affective states in valence, arousal, and dominance. The main idea is that human emotions can be quantified by measuring their degree of emotions. To the best of our knowledge, this is the first step in this direction, and we have proposed and successfully implemented it to induce feelings in robots and games.