P. Fung and K. McKeown. A technical word- and term-translation aid using noisy parallel corpora across language groups. Machine Translation, 12:53--87, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....by fusing different linguistic hypotheses. A preliminary evaluation is also reported. Introduction Most of the current corpus based machine translation systems rely on statistical methods to extract bilingual dictionaries. While these methods, such as [Kay and Rscheisen 1993] K vec and DK vec [Fung and McKeown 1994, 1997], and [Brown 1997] presume little or no prior knowledge of source languages, they require significant amount of parallel text to build an accurate bilingual dictionary [Jones and Somers 1995] Somers and Ward 1996] Haruno and Yamazaki 1996] The requirement makes these approaches less desirable ....

Fung, P. and KcKeown, K. (1997) A Technical Word- and Term-Translation Aid Using Noisy Parallel Corpora Across Language Groups. Machine Translation, Vol. 12, Nos. 1-2, pp. 53-87.

....(Melamed 1997a) 5. Limit the evaluation to certain kinds of type links based on the purpose of the evaluation, for example terminology, content words, etc. In some work presented in the literature, the explicit goal has been to extract bilingual dictionaries (e.g. Klavans Tzoukermann 1990 and Fung McKeown 1996). Bilingual terminology extraction can also be seen as a kind of specialization of bilingual extraction systems (e.g. Dagan Church 1994) As mentioned earlier in section 1.1, several methods can be used to evaluate an automatically created dictionary, for example by automatic comparison with a ....

.... As far as we can tell, there are no standard way of calculating scores in terms of precision and recall for extracted bilingual lexicons, but an alternative evaluation method could be a more pragmatic and practical approach, similar to the solutions suggested by Dagan Church (1994) and Fung McKeown (1996). Both adopt a way to measure the increase in efficiency that can be observed when translators are using a particular machineextracted dictionary. The translators who tested the dictionaries extracted by Fung and McKeown, for example, increased the number of correct term translations by 47 per ....

Fung, P. & K. McKeown (1996) A Technical Word and Term Translation Aid using Noisy Parallel Corpora Across Language Groups. The Machine Translation Journal, Special Issue on New Tools for Human Translators, pp. 53-87.

.... use sentence aligned parallel texts to further compile a bilingual lexicon of technical words or terms using similarity measures on bilingual lexical pairs [21, 25, 29] Yet others focus on translating phrases or terms which consist of multiple words [6, 25, 29] The main inspiration for our work [10, 14], to be described in the following section, comes from [21] who propose using word occurrences patterns and average mutual information and t scores to find word correspondences as an alternative to the IBM word alignment model. Given any pair of bilingual words, Lecture Notes in Computer Science ....

....parallel corpora Our work in bilingual lexicon extraction is motivated by the need to handle parallel corpora which do not have clear sentence boundaries, which contain many insertion or deletion noise, which consist of language pairs across families. These corpora are difficult to align. DKvec [10, 13, 14] is an algorithm which, instead of looking at the position vector of a word, looks at the arrival distance vector of the word and compares it with that of target words. An arrival distance vector is a vector whose values are the positional differences between two successive occurrences of the word ....

Pascale Fung and Kathleen McKeown. A technical word and term translation aid using noisy parallel corpora across language groups. Machine Translation, pages 53--87, 1996.

No context found.

P. Fung and K. McKeown. A technical word- and term-translation aid using noisy parallel corpora across language groups. Machine Translation, 12:53--87, 1997.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC