A Computational Model of Collaborative Discourse

This paper presents a model of collaborative discourse which integrates problem-solving, discourse, dialogue initiative and connict resolution. Experimental evidence obtained from an implementation of this theory demonstrates the eeectiveness of dialogue initiative and connict resolution within the model. 1 Motivation Two recently implemented voice dialogue systems utilize the principle that discourse is a by-product of cooperative problem-solving 11, 14]. The computer participant in both of these systems proceeds by attempting to prove or solve some goal, and when the system is blocked, it has the option of using dialogue to obtain the necessary information in order to continue its proof. This dialogue mechanism has been described as the missing axiom theory of discourse 10]. This paper extends this theory to develop a uniied model of collabora-tive discourse within the missing axiom theory framework which provides for collaborative behaviors not obtainable from previous versions of the missing axiom theory: 1) automatic adjustment of levels of initiative between participants during a discourse and 2) generation of certain important classes of statements (negotiation and summaries). Furthermore, this paper presents experimental data obtained through an implementation of this model which illustrates an increase in eeciency due to these additional dialogue behaviors.