DEPFIX: A System for Automatic Correction of Czech MT Outputs

Proceedings of the 7th Workshop on Statistical Machine Translation, pages 362–368, Montr´eal, Canada, June 7-8, 2012. c⃝2012 Association for Computational Linguistics DEPFIX: A System for Automatic Correction of Czech MT Outputs∗ Rudolf Rosa, David Mareˇcek and Ondˇrej Duˇsek Charles University in Prague, Faculty of Mathematics and Physics Institute of Formal and Applied Linguistics Malostransk´e n´amˇest´ı 25, Prague {rosa,marecek,odusek}@ufal.mff.cuni.cz Abstract We present an improved version of DEPFIX (Mareˇcek et al., 2011), a system for auto- matic rule-based post-processing of English- to-Czech MT outputs designed to increase their fluency. We enhanced the rule set used by the original DEPFIX system and measured the performance of the individual rules. We also modified the dependency parser of McDonald et al. (2005) in two ways to adjust it for the parsing of MT outputs. We show that our system is able to improve the quality of the state-of-the-art MT systems. 1 Introduction The today’s outputs of Machine Translation (MT) often contain serious grammatical errors. This is particularly apparent in statistical MT systems (SMT), which do not employ structural linguistic rules. These systems have been dominating the area in the recent years (Callison-Burch et al., 2011). Such errors make the translated text less fluent and may even lead to unintelligibility or misleading statements. The problem is more evident in lan- guages with rich morphology, such as Czech, where morphological agreement is of a relatively high im- portance for the interpretation of syntactic relations. The DEPFIX system (Mareˇcek et al., 2011) at- tempts to correct some of the frequent SMT sys- ∗This research has been supported by the European Union Seventh Framework Programme (FP7) under grant agree- ment n◦ 247762 (Faust), and by the grants GAUK116310, GA201/09/H057 (Res-Informatica), and LH12093. tems’ errors in English-to-Czech translations.1 It an- alyzes the target sentence (the SMT output in Czech language) using a morphological tagger and a de- pendency parser and attempts to correct it by apply- ing several rules which enforce consistency with the Czech grammar. Most of the rules use the source sentence (the SMT input in English language) as a source of information about the sentence structure. The source sentence is also tagged and parsed, and word-to-word alignment with the target sentence is determined. In this paper, we present DEPFIX 2012, an im- proved version of the original DEPFIX 2011 system. It makes use of a new parser, described briefly in Section 3, which is adapted to handle the generally ungrammatical target sentences better. We have also enhanced the set of grammar correction rules, for which we give a detailed description in Section 4. Section 5 gives an account of the experiments per- formed to evaluate the DEPFIX 2012 system and compare it to DEPFIX 2011. Section 6 then con- cludes the paper. 2 Related Work Our approach can be regarded as converse to the more common way of using an SMT system to auto- matically post-edit the output of a rule-based transla- tion system, as described e.g. in (Simard et al., 2007) or (Lagarda et al., 2009). …

Content truncated to 500 words