A web-based platform for annotating sentiment-related phenomena in human-agent conversations

Abstract

This paper introduces a web-based platform dedicated to the annotation of sentiment-related phenomena in human-agent conversations. The platform focuses on verbal content and deliberately sets aside non-verbal features. It is designed for managing two dialogue features: adjacency pair and conversation progression. Two annotation tasks are considered: (i) the detection of sentiment expressions, (ii) the ranking of user’s preferences. These two tasks focus on a set of specific targets. With this demonstration, we aim to introduce this platform to a large scientific audience and to get feedback for future improvements. Our long-term goal is to make the platform available as open-source tool.