PROJECT TITLE :

Sentiment Analysis: From Opinion Mining to Human-Agent Interaction

ABSTRACT:

The opinion mining and human-agent interaction communities are currently addressing sentiment analysis from completely different perspectives that comprise, on the one hand, disparate sentiment-related phenomena and computational representations, and on the opposite hand, totally different detection and dialog management methods. During this paper we tend to establish and discuss the growing opportunities for cross-disciplinary work that may increase individual advances. Sentiment/opinion detection strategies employed in human-agent interaction are indeed rare and, after they are used, they're not completely different from those utilized in opinion mining and consequently not designed for socio-affective interactions (timing constraint of the interaction, sentiment analysis as an input and an output of interaction strategies). To support our claims, we present a comparative state-of-the-art that analyzes the sentiment-related phenomena and therefore the sentiment detection ways utilized in each communities and makes an summary of the goals of socio-affective human-agent methods. We tend to propose then different potentialities for mutual profit, specifying many analysis tracks and discussing the open questions and prospects. To show the feasibility of the overall tips proposed we tend to also approach them from a particular perspective by applying them to the case of the Greta embodied conversational agents platform and discuss the means they will be used to form a more significative sentiment analysis for human-agent interactions in 2 completely different use cases: job interviews and dialogs with museum visitors.