Microplanning with Communicative Intentions: The SPUD System

Abstract

The process of microplanning encompasses a range of problems in Natural Language Generation (NLG), such as referring expression generation, lexical choice, and aggregation, problems in which a generator must bridge underlying domain-specific representations and general linguistic representations. In this paper, we describe a uniform approach to microplanning based on declarative representations of a generator's communicative intent. These representations describe the RESULTS of NLG: communicative intent associates the concrete linguistic structure planned by the generator with inferences that show how the meaning of that structure communicates needed information about some application domain in the current discourse context. Our approach, implemented in the SPUD (sentence planning using description) microplanner, uses the lexicalized treeadjoining grammar formalism (LTAG) to connect structure to meaning and uses modal logic programming to connect meaning to context. At the same time, communicative intent representations provide a RESOURCE for the PROCESS of NLG. Using representations of communicative intent, a generator can augment the syntax, semantics and pragmatics of an incomplete sentence simultaneously, and can assess its progress on the various problems of microplanning incrementally. The declarative formulation of communicative intent translates into a well-defined methodology for designing grammatical and conceptual resources which the generator can use to achieve desired microplanning behavior in a specified domain.

Pursuing and demonstrating understanding in dialogue

“…Our approach to grounding elaborates the model of NLG as problem solving from the SPUD generator [Stone et al, 2003]. SPUD solves problems of contributing specific new information to interlocutors against the backdrop of a determinate common ground.…”

Data-based analysis of speech and gesture: the Bielefeld Speech and Gesture Alignment corpus (SaGA) and its applications

“…Since the topic does not always coincide with an already introduced discourse referent [40, p. 265] we additionally distinguish the information foci given and new. Borrowing the terminology of Stone et al [69], information foci are classified according to the information states private or shared.…”

Towards Conversational Agents That Attend to and Adapt to Communicative User Feedback

“…Billie's behaviour planning component contains a novel natural language microplanner based on the SPUD framework [23], which has been extended to take the attributed listener state into account while generating utterances chunk by chunk. To this end, the linguistic constructions used by the microplanner are annotated with information about their verbosity and then chosen according to the ALS's meta-variable 'difficulty of perception', leading to utterances that are more verbose, i.e., using more words to express the same idea, if the user has problems perceiving the agent.…”