While it is has often been observed that the product of translation is somehow different than non-translated text, scholars have emphasized two distinct bases for such differences. Some have noted interference from the source language spilling over into translation in a source-language-specific way, while others have noted general effects of the process of translation that are independent of source language. Using a series of text categorization experiments, we show that both these effects exist and that, moreover, there is a continuum between them. There are many effects of translation that are consistent among texts translated from a given source language, some of which are consistent even among texts translated from families of source languages. Significantly, we find that even for widely unrelated source languages and multiple genres, differences between translated texts and non-translated texts are sufficient for a learned classifier to accurately determine if a given text is translated or original. 1

We investigate the differences between language models compiled from original target-language texts and those compiled from texts manually translated to the target language. Corroborating established observations of Translation Studies, we demonstrate that the latter are significantly better predictors of translated sentences than the former, and hence fit the reference set better. Furthermore, translated texts yield better language models for statistical machine translation than original texts. 1

Abstract not found

We revisit the one sense per discourse hypothesis of Gale et al. in the context of machine translation. Since a given sense can be lexicalized differently in translation, do we observe one translation per discourse? Analysis of manual translations reveals that the hypothesis still holds when using translations in parallel text as sense annotation, thus confirming that translational differences represent useful sense distinctions. Analysis of Statistical Machine Translation (SMT) output showed that despite ignoring document structure, the one translation per discourse hypothesis is strongly supported in part because of the low variability in SMT lexical choice. More interestingly, cases where the hypothesis does not hold can reveal lexical choice errors. A preliminary study showed that enforcing the one translation per discourse constraint in SMT can potentially improve translation quality, and that SMT systems might benefit from translating sentences within their entire document context. 1

(but not limited to) processing of Hebrew. Some projects require previous knowledge of computational linguistics but some assume no previous background. All projects (except one) involve programming: the end result is a relatively large-scale, well-documented and efficient software package. Some of the projects may involve also some research (e.g., reading a research paper and implementing its ideas). 2 Administration Projects are to be implemented by groups of at most two students. All systems will be presented at the end of the semester for a final demo. A coordination meeting is planned for Wednesday, June 2nd; all work must be completed by Tuesday, August 31st. A project presentation meeting will be held on Wednesday, September 1st. The programming language must be portable enough to be usable on a variety of platforms; Python is recommended, C++, Perl or Java will be tolerated, if you have a different language in mind discuss it with the instructor. Most projects will have to be executed in a Linux environment due to dependencies on external packages. Grading will be based on comprehension of the problem, quality of the implementation and quality of the documentation. In particular, the final grade will be based on: Comprehension of the problem (and the accompanying paper(s), where applicable); Full implementation of a working solution; Presentation of a final working system; Comprehensive documentation. Computational Linguistics Group

new text statistical measure and its application to stylometry

(but not limited to) processing of Hebrew. Some projects require previous knowledge of computational linguistics but some assume no previous background. All projects (except one) involve programming: the end result is a relatively large-scale, well-documented and efficient software package. Some of the projects may involve also some research (e.g., reading a research paper and implementing its ideas). 2

Translation models used for statistical machine translation are compiled from parallel corpora; such corpora are manually translated, but the direction of translation is usually unknown, and is consequently ignored. However, much research in Translation Studies indicates that the direction of translation matters, as translated language (translationese) has many unique properties. Specifically, phrase tables constructed from parallel corpora translated in the same direction as the translation task perform better than ones constructed from corpora translated in the opposite direction. We reconfirm that this is indeed the case, but emphasize the importance of using also texts translated in the ‘wrong ’ direction. We take advantage of information pertaining to the direction of translation in constructing phrase tables, by adapting the translation model to the special properties of translationese. We define entropybased measures that estimate the correspondence of target-language phrases to translationese, thereby eliminating the need to annotate the parallel corpus with information pertaining to the direction of translation. We show that incorporating these measures as features in the phrase tables of statistical machine translation systems results in consistent, statistically significant improvement in the quality of the translation.