A self-learning method of parallel texts alignment

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper describes a language independent method for alignment of parallel texts that re-uses acquired knowledge. The system extracts word translation equivalents and re-uses them as correspondence points in order to enhance the alignment of parallel texts. Points that may cause misalignment are filtered using confidence bands of linear regression analysis instead of heuristics, which are not theoretically reliable. Homographs bootstrap the alignment process so as to build the primary word translation lexicon. At each step, the previously acquired lexicon is re-used so as to repeatedly make finer-grained alignments and produce more reliable translation lexicons.

Original languageEnglish
Title of host publicationEnvisioning Machine Translation in the Information Future
Subtitle of host publication4th Conference of the Association for Machine Translation in the Americas, AMTA 2000, Proceedings
EditorsJohn S. White
Place of PublicationBerlin
PublisherSpringer
Pages30-39
Number of pages10
ISBN (Electronic)978-3-540-39965-0
ISBN (Print)3540411178, 978-3-540-41117-8
DOIs
Publication statusPublished - 2000
Event4th Conference of the Association for Machine Translation in the Americas, AMTA 2000 - Cuernavaca, Mexico
Duration: 10 Oct 200014 Oct 2000

Publication series

NameLecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)
Volume1934
ISSN (Print)0302-9743

Conference

Conference4th Conference of the Association for Machine Translation in the Americas, AMTA 2000
Country/TerritoryMexico
CityCuernavaca
Period10/10/0014/10/00

Fingerprint

Dive into the research topics of 'A self-learning method of parallel texts alignment'. Together they form a unique fingerprint.

Cite this