Papers by Pierre Isabelle
In this paper, we show how a large bilingual English-French parallel corpus can be brought to bea... more In this paper, we show how a large bilingual English-French parallel corpus can be brought to bear in terminology search. First, we demonstrate that the coverage of available corpora has become substantially more extensive than that of mainstream term banks. One potential drawback in searching large unstructured corpora is that large numbers of search results may need to be examined before finding a relevant match. We argue that this problem can be alleviated by contextualizing the search process: instead of looking up isolated terms one searches for terms appearing in a context that is similar to that of the term to be translated. We present an experiment on contextbased re-ranking and report highly positive results. We conclude that translators will increasingly rely on very large scale corpora for searching term equivalents.
Computational Linguistics, 1985
Upon the completion of its highly successful TAUM-METEO machine translation system, the TAUM grou... more Upon the completion of its highly successful TAUM-METEO machine translation system, the TAUM group undertook the construction of TAUM-AVIATION, an experimental system for English to French translation in the sublanguage of technical maintenance manuals. A detailed description of the resulting prototype is offered. In particular, the paper includes: a) some figures on the size of the system; b) a description of the underlying translation model (indirect approach, analysis/transfer/synthesis scheme); c) a presentation of the basic computational techniques (use of a specialized high-level metalanguage for each linguistic component); and d) some results on the evaluation of the prototype.
Making human translators more productive • (Koehn & Haddow): three options for translators: 1) su... more Making human translators more productive • (Koehn & Haddow): three options for translators: 1) suggestions for sentence completion; 2) word & phrase translation options, and 3) post-editing of MT. Fr → En user study looks at productivity impact and users' impressions.
Meta: Translators' Journal, 1981
We present a progress report on our research on nominal compounds (NC's). Recent approaches to th... more We present a progress report on our research on nominal compounds (NC's). Recent approaches to this probiem in linguistics and natural ianguage processing (NLP) are reviewed and criticized. We argue that the notion of "roie nominal", which is at the interface of linguistic and extraiinguistic knowledge, is crucial for characterizing NC'e as weII as other Iinguistic phenomena. We examine a number of constraints on the semantic interpretation ruies for NC's. Proposals are made that shouid improve the capability of NLP systems to deaI with NC's.
Meeting of the Association for Computational Linguistics, Jul 6, 2002
This year's meeting is special. First, it the first ACL meeting to be hosted jointly with our... more This year's meeting is special. First, it the first ACL meeting to be hosted jointly with our North-American chapter, the NAACL. Second, it is also the ACL's 40th anniversary meeting: our association was created 13 June 1962, and its first meeting took place 25-26 August 1963 in Denver.Our field has expanded quite a bit since then, and so have ACL meetings. This year, our main conference program includes some 65 different refereed papers, covering most active research areas in computational linguistics. And our extended program is complementing that with 19 system demonstrations, 4 tutorials, 12 Student Research papers, and no less than 11 different pre- and post-conference workshops presenting a total of about 150 papers.
Cahier de linguistique, 1978
Machine Translation, 1993
We argue that the conventional approach to Interactive Machine ~Ih-anslation is not the best way ... more We argue that the conventional approach to Interactive Machine ~Ih-anslation is not the best way to provide assistance to skilled translators, and propose an alternative whose central feature is the use of the target text as a medium of interaction. We describe an automatic wordcompletion system intended to serve as a vehicle for exploring the feasibility of this new approach, and give results in terms of keystrokes saved in a test corpus.
Meeting of the Association for Computational Linguistics, 1998
There is an increasing need for document search mechanisms capable of matching a natural language... more There is an increasing need for document search mechanisms capable of matching a natural language query with documents written in a different language. Recently, we conducted several experiments aimed at comparing various methods of incorporating a cross-linguistic capability to existing information retrieval (IR) systems. Our results indicate that translating queries with off-theshelf machine translation systems can result in relatively good performance. But the results also indicate that other methods can perfonn even better. More specifically, we tested a probabilistic translation model of the kind proposed by Brown & al. [2]. The parameters of that system had been estimated automatically on a different, unrelated, corpus of parallel texts. After we augmented it with a small bilingual dictionary, this probabilistic translation model outperformed machine translation systems on our cross-language IR task.
arXiv (Cornell University), Sep 28, 1994
Professional translators often dictate their translations orally and have them typed afterwards. ... more Professional translators often dictate their translations orally and have them typed afterwards. The TransTalk project aims at automating the second part of this process. Its originality as a dictation system lies in the fact that both the acoustic signal produced by the translator and the source text under translation are made available to the system. Probable translations of the source text can be predicted and these predictions used to help the speech recognition system in its lexical choices. We present the results of the first prototype, which show a marked improvement in the performance of the speech recognition task when translation predictions are taken into account.
Uploads
Papers by Pierre Isabelle