On the Place of Words In the Generation Process

Natural Language Generation in Artificial Intelligence and Computational Linguistics

“…Schemata have been successfully employed in several systems to perform tasks such as describing and comparing entities in a database , providing corrective responses to a user's object-related misconceptions (McCoy, 1989), and describing complex physical objects (Paris, 1988). Schemata have been successfully employed in several systems to perform tasks such as describing and comparing entities in a database , providing corrective responses to a user's object-related misconceptions (McCoy, 1989), and describing complex physical objects (Paris, 1988).…”

“…Such decision making is done under constraints and the ability to clearly and concisely represent constraints is important (McKeown and Paris, 1987). Such decision making is done under constraints and the ability to clearly and concisely represent constraints is important (McKeown and Paris, 1987).…”

“…Much recent research has been devoted to demonstrating how the quality of responses generated can be improved given a detailed model of the user -including what the user knows about the domain, how information should be presented to that user, and so forth (Appelt, 1985;McCoy, 1989;Paris, 1988;Kass and Finin, 1988). Moreover, if a system has only one opportunity to produce a text that achieves the speaker's goals without overor under-informing, boring or confusing the listener, then that system must have an enormous amount of detailed knowledge about the listener.…”

“…At any point in a discourse the focused knowledge represents the knowledge that a natural language generation system could reasonably be expected to include next in its interaction with the user. For examples see (McKeown, 1985a;McCoy, 1985;Paris and McKeown, 1987;Mann and Thompson, 1987;Hovy, 1988a;Moore and Swartout, 1990) among others. Moreover, the shift in focus that it chooses cannot be random, but instead must be a shift that can be readily followed by the human conversational participant.…”

Lexical Semantics and Knowledge Representation in Multilingual Text Generation

“…(Section 2.4) Quite often, researchers have lamented that the problem of word choice has not received sufficient attention, e.g., [Marcus 1987;Nirenburg and Nirenburg 1988;McDonald 1991]-many language generators assume that for every concept in the input expression there is exactly one associated word. Yet, when lexicalization is indeed seen as a matter of choice, factors determining the differences between lexical items need to be found, and taking at least some of them into consideration can enhance the expressiveness of a generator considerably.…”