1991 | OriginalPaper | Buchkapitel
Focus of attention: Constraining what can be said next
verfasst von : Kathleen F. McCoy, Jeannette Cheng
Erschienen in: Natural Language Generation in Artificial Intelligence and Computational Linguistics
Verlag: Springer US
Enthalten in: Professional Book Archive
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
During the course of a coherent conversation or discourse, the participants their attention on some subset of their knowledge. As the discourse goes on, this focused subset may change — it may grow to include more knowledge, narrow down to include just a subset of what it originally contained, or shift (either temporarily or permanently) to a new area of the participants’ knowledge base. In this work we investigate the nature of this focusing during a discourse and its effect on a natural language generation system. Our research examines how the focused knowledge is tracked during human discourse and how changes in the focused set are marked by the human conversational partners. We hypothesize that there are several different kinds of focusing going on in discourse, and attempt to provide a unified account which can handle each. This resulting knowledge will be crucial for a generation system in deciding what to say next and in deciding how to appropriately mark unexpected changes in focus.