A Formal Mechanism to Process Natural Language: Grammatical Number Recognition
Date
1998
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Producer
Director
Performer
Choreographer
Costume Designer
Music
Videographer
Lighting Designer
Set Designer
Crew Member
Funder
Rehearsal Director
Concert Coordinator
Advisor
Moderator
Panelist
Alternative Title
Department
Swarthmore College. Dept. of Linguistics
Type
Thesis (B.A.)
Original Format
Running Time
File Format
Place of Publication
Date Span
Copyright Date
Award
Language
en_US
Note
Table of Contents
Terms of Use
Full copyright to this work is retained by the student author. It may only be used for non-commercial, research, and educational purposes. All other uses are restricted.
Rights Holder
Access Restrictions
Terms of Use
Tripod URL
Identifier
Abstract
The purpose of this research is to attempt to create a grammar or mechanism
that will be able to generate or comprehend grammatical number strings in
Standard American English while either not generating or not accepting
ungrammatical number strings. Clearly, the human brain is capable of performing
this task of weeding out any ungrammatical strings and comprehending and
generating legal strings. For example, when a native speaker of English is presented
with the string "one hundred forty four thousand three hundred twelve" it is
conceivable that the speaker has never encountered this number before, never
spoken it, never heard it, and never read it. However, the hearer readily recognizes
this string as a legal number, a possible combination of components in the English
language. At the same time, the speaker will never generate or accept as
grammatical a string such as "forty hundred three thousand one thousand." Every
individual word in that string is a word of English, and a number term, but even
though the speaker has never been taught that that specific string is not a number,
the speaker intuitively recognizes it as nonsense. The questions that I seek to
answer are: is it possible to mimic the rules in the human processing of these
strings that allow only legal strings to be recognized as legitimate numbers, and
what mechanisms are most efficient and effective in simulating human number
comprehension.
In this research, we shall be concerned entirely with number strings used to
describe real quantities, the numbers we use to count things. We will not be
concerned with numbers such as "nineteen ninety seven" (as in the year 1997), nor
will we be concerned with numbers that do not correspond to real quantities, such
as "a million billion trillion."
The idea for this project stems from work I undertook at the Institute for
Research in Cognitive Science at the University of Pennsylvania, in the summer of
1997. The models used will be pencil and paper models, not actual working devices.