Architecture of a Socio-Emotional Conversational Agent Capable of Identifying Intentions within a Socratic Dialogue

Abstract

This article presents an architecture for developing a socio-emotional conversational agent to identify intentions within Socratic dialogues, using a combination of speech act theory and sentiment analysis. This proposal seeks to create a socio-emotional conversational agent capable of addressing the issue of how emotions associated with personality affect the recognition of intentions of expressions within the educational environment. To identify the appropriate intention, this approach considers the influence of personality, the polarity of the words used by the user, and the context in which the dialogue is situated. With this, it is expected to accurately identify communicative intentions, facilitating a natural and empathetic interaction between intelligent virtual agents and users, contributing to the field of artificial intelligence applied to education.