Publikation

A Shallow Formalism for Defining Personalized Text

Stephan Busemann

In: Workshop Professionelle Erstellung von Papier- und Online-Dokumenten: Perspektiven für die automatische Textgenerierung. 22nd Annual German Conference on Artificial Intelligence (KI '98), 15.-17. September. German Conference on Artificial Intelligence (KI) 1998.

Abstrakt

In this workshop note, we sketch techniques suited to generate personalized text within the shallow verbalization system TG/2. In a grammar-engineering phase, alternative formulations of messages corresponding to different user types are encoded into the generation grammar. A set of parameters denoting properties of verbalization alternatives is associated with the respective grammar rules. The user provides to the system the parameter values she desires. During generation, the rules matching the criteria best are preferred, leading to personalized text.

Weitere Links

busemann98b.dvi.gz (gz, 8 KB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence