Generating Questions: An Inclusive Characterization and a Dialogue-based Application
Paul Piwek, Helmut Prendinger, Hugo Hernault, Mitsuru Ishizuka · 2008
In this paper, we make a case for the intrinsic richness of the Question Generation (QG) task. We examine QG in terms of its input, output and the relation between the two. We then describe a specific approach to QG which depends on -abstraction to formalize the relation between questions and answers. This allows us to make fine-grained distinctions between different types of questions. We have applied this approach to the generation of dialogue from monologue, exploiting the coherence relations underlying the monologue.