In this paper, we present a social model for software agent conversations for action based on social commitments and their negotiation. We depart from the premises that conversations are cornerstone to support autonomous and heterogeneous agents, and that conversational coherence can be supported through public definitions of speech act and compositional semantics. We specify a unified social model for conversations in which speech act semantics is an emergent product of identity, conversational use, and potential consequences, and where conversational composition is guided by rules of conversational use and their application to the state of conversation instances. Lastly, we show the effectiveness of this novel approach by formally describing the evolution of a simple conversation for action.
CITATION STYLE
Flores, R. A., & Kremer, R. C. (2002). Bringing coherence to agent conversations. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2222, pp. 50–67). Springer Verlag. https://doi.org/10.1007/3-540-70657-7_4
Mendeley helps you to discover research relevant for your work.