Generating bilingual lexical equivalences from parallel texts

Boutsis, S; Piperidis, S; Demiros, I

dc.contributor.author	Boutsis, S	en
dc.contributor.author	Piperidis, S	en
dc.contributor.author	Demiros, I	en
dc.date.accessioned	2014-03-01T01:48:53Z
dc.date.available	2014-03-01T01:48:53Z
dc.date.issued	1999	en
dc.identifier.issn	0883-9514	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/25627
dc.subject.classification	Computer Science, Artificial Intelligence	en
dc.subject.classification	Engineering, Electrical & Electronic	en
dc.title	Generating bilingual lexical equivalences from parallel texts	en
heal.type	journalArticle	en
heal.language	English	en
heal.publicationDate	1999	en
heal.abstract	This paper describes the application of artificial intelligence methods fort the automatic extraction of translation equivalences from bilingual parallel texts. In an attempt to implement as language independent a method as possible, the application's methodology features statistical inductive techniques coupled with symbolic processing techniques catering for the analysis of specific language phenomena. The method presupposes parallel texts and identifies translational equivalences at the word or multiword unit level for. those cases that such an equivalence holds the. Pal allel texts are first st aligned at the sentence level and grammatically analyzed. Noun phr ase grammars extract noun phrases, and statistical evaluation yields the most coherent multiword units on either side. Translation candidates of word ol multiword units are evaluated by a similarity metric defined by the co-occurrence frequency and independent frequencies of the units. The method has been tested on an English-Greek corpus consisting of texts relevant to software systems, yielding similar to 94% accurate translations.	en
heal.publisher	TAYLOR & FRANCIS INC	en
heal.journalName	APPLIED ARTIFICIAL INTELLIGENCE	en
dc.identifier.isi	ISI:000082008100002	en
dc.identifier.volume	13	en
dc.identifier.issue	6	en
dc.identifier.spage	583	en
dc.identifier.epage	606	en