We report on the benefits achieved by using ontologies in the context of a fully implemented conversational system that allows for a real-time rich communication between primarily 10 to 18 years old human users and a 3D graphical character through spontaneous speech and gesture. In this paper, we focus on the categorization of ontological resources into domain independent and domain specific components in the effort of both augmenting the agent's conversational capabilities and enhancing system's reusability across conversational domains. We also present a novel method of exploiting the existing ontological resources along with Google directory categorization for a semi-automatic understanding of user utterance on general purpose topics like e.g. movies and games. © Springer-Verlag Berlin Heidelberg 2007.
CITATION STYLE
Mehta, M., & Corradini, A. (2007). Developing a conversational agent using ontologies. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4552 LNCS, pp. 154–164). Springer Verlag. https://doi.org/10.1007/978-3-540-73110-8_17
Mendeley helps you to discover research relevant for your work.