Locating redundant segments using a translation memory, Transit
نویسنده
چکیده
The purpose of this work was to find a method for locating redundant information between chapters of the same manual, by using the translation memory Transit provided by STAR group. A filter for comparing the similarity of sentences, Fuzzy filter, was set to assemble sentences with >50% recemblance and the results were saved. Concordance search—based on the result from the previous step and the terminology lexicon—was set to retreive segments with 80% similarity and to display 300 matches. The results showed that the number of segments with similarities in the reference material that also contained terms were up to 3% (WSM PGRT-TI). The results are lower than STAR translitera’s previous method for locating overlapping information, which however was not term dependent. The results open for discussion on how to improve the quality and impact of allowed and forbidden terms respectively.
منابع مشابه
MT-enhanced fuzzy matching with Transit NXT and STAR Moses
The first step was integration at a project-handling level: As with standard TMS, the text is first pretranslated using validated translations from the translation memory. The remaining “deltas” (new or changed segments) are sent to the MT engine for translation. This means that the translator is offered two types of suggestion for “deltas”: Fuzzy matches from the translation memory and machine...
متن کاملTLB Performance in Multiprocessors by Patricia
This paper compares the performance, in highly-parallel shared-memory multiprocessors, of locating translation-lookaside buffers (TLBs) at processors with that of locating TLBs at memory. Our performance comparison is based on results of trace-driven simulations of multiprocessors with logN-stage networks interconnecting N processors and N memory modules. For the systems and workloads studied, ...
متن کاملEvaluating the Operational Benefit of Using Machine Translation Output as Translation Memory Input
Following the guidelines for MT evaluation proposed in the ISLE taxonomy, this paper presents considerations and procedures for evaluating the integration of machine-translated segments into a larger translation workflow with Translation Memory (TM) systems. The scenario here focuses on the software localisation industry, which already uses TM systems and looks to further streamline the overall...
متن کاملSelecting Syntactic, Non-redundant Segments in Active Learning for Machine Translation
Active learning is a framework that makes it possible to efficiently train statistical models by selecting informative examples from a pool of unlabeled data. Previous work has found this framework effective for machine translation (MT), making it possible to train better translation models with less effort, particularly when annotators translate short phrases instead of full sentences. However...
متن کاملProfessionelles Übersetzen mit STAR Transit
Transit (Translate it) ist ein professionelles Translation-Memory (TM) System entwickelt auf der Basis der langjährigen praktischen Erfahrung von STAR, eines der größten und erfolgreichsten Übersetzungsunternehmens der Welt. Die Kunden von STAR kommen aus der Automobilindustrie, der Informationstechnologie und dem Bereich Maschinenbau. Transit, das seit 1994 kommerziell als Produkt vertrieben w...
متن کامل