Finetuning NLG Through Experiments with Human Subjects: The Case of Vague Descriptions

This discussion paper describes a sequence of experiments with human subjects aimed at finding out how an nlg system should choose between the different forms of a gradable adjective. This case study highlights some general questions that one faces when trying to base nlg systems on empirical evidence: one question is what task to set a subject so as to obtain the most useful information about that subject, another question has to do with differences between subjects.

• Type: Conference
• Year: 2004
• Where: INLG
• Authors: Kees van Deemter
• Keywords: Gradable Adjective, Human Subjects, INLG 2004, Natural Language Processing, Nlg Systems