Charles B. Callaway and James Lester
In recent years, the natural language generation community has begun
to mature rapidly and produce sophisticated off-the-shelf surface
realizers. A parallel development in the knowledge representation
community has been the emergence of large-scale knowledge bases that
house tens of thousands of facts encoded in expressive
representational languages. Because of the richness of their
representations and the sheer volume of their formally encoded
knowledge, these knowledge bases offer the promise of significantly
improving the quality of natural language generation. However, the
representational complexity, scale, and task-independence of these
knowledge bases pose great challenges to natural language generators.
We have designed, implemented, and empirically evaluated FARE, a
functional realization system that exploits message
specifications drawn from large-scale knowledge bases to create
functional descriptions, which are expressions that encode both
functional information (case assignment) and structural information
(phrasal constituent embeddings). Given a message specification,
FARE exploits lexical and grammatical annotations on knowledge base
objects to construct functional descriptions, which are then converted
to text by a surface generator. Two empirical studies---one with an
explanation generator and one with a qualitative model
builder---suggest that FARE is robust, efficient, expressive,
and appropriate for a broad range of applications.