Syllables and the M Language : improving unknown word guessing
Name
399645095-MIT.pdf
Description
Full printable version
Size
3.01 MB
Format
Adobe PDF
Checksum (MD5)
f60d25fcff0ac475be1c3dcaae8a94bd
Author(s)
Jacokes, M. Brian (Michael Brian)
Advisor(s)
David L. Brock.
Alternative Title
Unknown word guessing in a semantic data language
Date Issued
2008
Publisher
Massachusetts Institute of Technology
Abstract
Despite the huge amount of computer data that exists today, the task of sharing information between organizations is still tackled largely on a case-by-case basis. The M Language is a data language that improves data sharing and interoperability by building a platform on top of XML and a semantic dictionary. Because the M Language is specifically designed for real-world data applications, it gives rise to several unique problems in natural language processing. I approach the problem of understanding unknown words by devising a novel heuristic for word decomposition called "probabilistic chunking," which achieves a 70% success rate in word syllabification and has potential applications in automatically decomposing words into morphemes. I also create algorithms which use probabilistic chunking to syllabify unknown words and thereby guess their parts of speech and semantic relations. This work contributes valuable methods to the areas of natural language processing and automatic data processing.
Description
Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2008.
Includes bibliographical references (p. 71-72).
Subjects
Electrical Engineering and Computer Science.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
M.I.T. theses are protected by
copyright. They may be viewed from this source for any purpose, but
reproduction or distribution in any format is prohibited without written
permission. See provided URL for inquiries about permission.
copyright. They may be viewed from this source for any purpose, but
reproduction or distribution in any format is prohibited without written
permission. See provided URL for inquiries about permission.
Persistent DSpace Link
