Semantic Valence Modeling: Emotion Recognition and Affective States in Context-Aware Systems

Abstract

Defining and describing a context requires knowledge (contextual information), while research is addressing a wider range of potential contextual information in a diverse range of domains the diversity of potential contextual information has not been effectively addressed. This paper considers the implementation of context and identifies emotion (more accurately emotional response) as a factor in the personalization of services as under-represented in the literature. We propose semantic valence modeling implemented in fuzzy rule-based systems as a potential solution to the implementation of emotional responses in context-aware systems. It is concluded that the effective implementation of emotional responses based on the posited approach will improve the accuracy of personalized service provision and additionally offers the potential to improve the levels of computational intelligence in context-aware domains and systems.