The Karlsruhe Institute for Technology Translation System for the ACL-WMT 2010

Proceedings of the Joint 5th Workshop on Statistical Machine Translation and MetricsMATR, pages 138–142, Uppsala, Sweden, 15-16 July 2010. c⃝2010 Association for Computational Linguistics The Karlsruhe Institute for Technology Translation System for the ACL-WMT 2010 Jan Niehues, Teresa Herrmann, Mohammed Mediani and Alex Waibel Karlsruhe Instiute of Technolgy Karlsruhe, Germany firstname.lastname@kit.edu Abstract This paper describes our phrase-based Sta- tistical Machine Translation (SMT) sys- tem for the WMT10 Translation Task. We submitted translations for the German to English and English to German transla- tion tasks. Compared to state-of-the-art phrase-based systems we preformed addi- tional preprocessing and used a discrim- inative word alignment approach. The word reordering was modeled using POS information and we extended the transla- tion model with additional features. 1 Introduction In this paper we describe the systems that we built for our participation in the Shared Trans- lation Task of the ACL 2010 Joint Fifth Work- shop on Statistical Machine Translation and Met- ricsMATR. Our translations are generated using a state-of-the-art phrase-based translation system and applying different extensions and modifica- tions including Discriminative Word Alignment, a POS-based reordering model and bilingual lan- guage models using POS and stem information. Depending on the source and target languages, the proposed models differ in their benefit for the translation task and also expose different correl- ative effects. The Sections 2 to 4 introduce the characteristics of the baseline system and the sup- plementary models. In Section 5 we present the performance of the system variants applying the different models and chose the systems used for creating the submissions for the English-German and German-English translation task. Section 6 draws conclusions and suggests directions for fu- ture work. 2 Baseline System The baseline systems for the translation directions German-English and English-German are both de- veloped using Discriminative Word Alignment (Niehues and Vogel, 2008) and the Moses Toolkit (Koehn et al., 2007) for extracting phrase pairs and generating the phrase table from the discrimi- native word alignments. The difficult reordering between German and English was modeled us- ing POS-based reordering rules. These rules were learned using a word-aligned parallel corpus. The POS tags for the reordering models are generated using the TreeTagger (Schmid, 1994) for all lan- guages. Translation is performed by the STTK Decoder (Vogel, 2003) and all systems are optimized to- wards BLEU using Minimum Error Rate Training as proposed in Venugopal et al. (2005). 2.1 Training, Development and Test Data We used the data provided for the WMT for train- ing, optimizing and testing our systems: Our training corpus consists of Europarl and News Commentary data, for optimization we use new- stest2008 as development set and newstest2009 as test set. The baseline language models are trained on the target language part of the Europarl and News Commentary corpora. Additional, bigger lan- guage models were trained on monolingual cor- pora. For both systems the News corpus was used while an English language model was also trained on the even bigger Gigaword corpus. 2.2 Preprocessing The training data was preprocessed before used for …

Content truncated to 500 words