A specification language for agglutinative aboriginal languages for use with finite-state spelling correction

Keating, Ron Andrew (2011) A specification language for agglutinative aboriginal languages for use with finite-state spelling correction. Masters thesis, Memorial University of Newfoundland.

[img] [English] PDF - Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

Download (9Mb)

Abstract

There are certain North American aboriginal languages which are in danger of becoming extinct. This is partially due to the younger generations learning more major world languages in order to communicate in an increasingly global society. Furthermore, these languages tend to have only developed writing systems relatively recently, and thus do not have a rich legacy of written works to help preserve them. In order to help alleviate this problem, certain tools are being developed to facilitate communication in those languages. One such tool that is expected to help is a digital spelling correction tool. Having such a tool will make it easier to create professional digital documents in those languages, as well as help educate speakers with regard to the proper spellings of words. -- To facilitate the creation of a spelling correction software tool, this thesis proposes a simplified specification language called FSCL. Linguists can use FSCL to specify the details of natural languages in a format that is easily readable by both humans and computers without having to sacrifice any relevant expressive power, thus allowing linguists and even language speakers to build and maintain a working model of the natural language in question. The syntactic and semantic details of FSCL are discussed, and an implementation built on existing finite-state natural language processing algorithms is detailed and tested with respect to a set of actual language data from one such aboriginal language, Innu.

Item Type: Thesis (Masters)
URI: http://research.library.mun.ca/id/eprint/9518
Item ID: 9518
Additional Information: Bibliography: leaves 87-92.
Department(s): Science, Faculty of > Computer Science
Date: 2011
Date Type: Submission
Library of Congress Subject Heading: Computational linguistics; Natural language processing (Computer science); Indigenous peoples--Languages--Orthography and spelling; Spelling errors--Software

Actions (login required)

View Item View Item

Downloads

Downloads per month over the past year

View more statistics