Wishful: A Discourse Planning System That Considers a User's Inferences

Abstract

The explicit consideration of an addressee's inferences during discourse planning affects the content and coherence of the generated discourse. The content of the discourse is affected because the consideration of an addressee's inferences may require the addition of information that addresses erroneous inferences or suggest the omission of easily inferred information. The coherence of the discourse is affected because the generated discourse should incorporate inferential relations that emerge from the consideration of an addressee's inferences. In this article, we describe a discourse planner that takes into consideration a user's inferences. In particular, we discuss its content planning and discourse structuring mechanisms. The content planning mechanism generates a set of rhetorical devices that achieves a given communicative goal. The discourse structuring mechanism takes into consideration schematic, inferential, and prescriptive relations between these rhetorical devices in order to organize them into a coherent sequence. Our discourse planner has been implemented in a system called WISHFUL that generates explanations about concepts in technical domains. We evaluated WISHFUL's performance to test the effect of considering a user's inferences during content planning and discourse organization and also to determine the suitability of the discourse produced by WISHFUL for different types of users. The results of our evaluation endorse the ideas embodied in the WISHFUL system.

Influence of context on users’ views about explanations for decision-tree predictions

“…In 1990-2000, explanations derived from knowledge bases were enhanced by addressing aspects of users' reasoning. Specifically, Zukerman and McConachy (1993) and Horacek (1997) considered potential inferences from explanations, omitting easily inferable information and addressing erroneous inferences; Korb et al (1997) took into account reasoning fallacies when explaining the reasoning of Bayesian Networks; and Stone (2000) generated instructions from which users could draw appropriate inferences about actions to take.…”

Applications of Rhetorical Structure Theory

“…The types of text generated include instruction manuals of different types (Rösner and Stede, 1992;Vander Linden and Martin, 1995;Wahlster et al, 1991), administrative forms (Not and Stock, 1994), user documentation (Hartley and Paris, 1997), descriptions of tourist sights (Krifka-Dobes and Novak, 1993), and descriptions of concepts (Zukerman and McConachy, 2001), which are all monologic discourse types. Interactive dialogue has also been addressed, mostly in instructional texts: explanatory discourse about electronic circuits (Cawsey, 1990), advisory dialogues (Moore and Paris, 1993), and dialogue interaction with a database (Fischer et al, 1994).…”

Inferences, suppositions and explanatory extensions in argument interpretation

“…The research reported by Joshi et al (1984), van Beek (1987 and Zukerman and McConachy (2001)) considers the addition of information to planned discourse to prevent or weaken a user's erroneous inferences from this discourse. Our mechanism for adding explanatory extensions to an interpretation prevents inferences that people find objectionable due to jumps in belief.…”