Sensible: L2 Translation Assistance by Emulating the Manual Post-Editing Process

Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), pages 541–545, Dublin, Ireland, August 23-24, 2014. Sensible: L2 Translation Assistance by Emulating the Manual Post-Editing Process Liling Tan, Anne-Kathrin Schumann, Jose M.M. Martinez and Francis Bond1 Universit¨at des Saarland / Campus, Saarbr¨ucken, Germany Nanyang Technological University1 / 14 Nanyang Drive, Singapore alvations@gmail.com, anne.schumann@mx.uni-saarland.de, j.martinez@mx.uni-saarland.de, bond@ieee.org Abstract This paper describes the Post-Editor Z sys- tem submitted to the L2 writing assis- tant task in SemEval-2014. The aim of task is to build a translation assistance system to translate untranslated sentence fragments. This is not unlike the task of post-editing where human translators improve machine-generated translations. Post-Editor Z emulates the manual pro- cess of post-editing by (i) crawling and ex- tracting parallel sentences that contain the untranslated fragments from a Web-based translation memory, (ii) extracting the pos- sible translations of the fragments indexed by the translation memory and (iii) apply- ing simple cosine-based sentence similar- ity to rank possible translations for the un- translated fragment. 1 Introduction In this paper, we present a collaborative submis- sion between Saarland University and Nanyang Technological University to the L2 Translation As- sistant task in SemEval-2014. Our team name is Sensible and the participating system is Post- Editor Z (PEZ). The L2 Translation Assistant task concerns the translation of an untranslated fragment from a par- tially translated sentence. For instance, given a sentence, “Ich konnte B¨arbel noch on the border in einen letzten S-Bahn-Zug nach Westberlin set- zen.”, the aim is to provide an appropriate transla- tion for the underline phrase, i.e. an der Grenze. The aim of the task is not unlike the task of post-editing where human translators correct er- rors provided by machine-generated translations. This work is licensed under a Creative Commons At- tribution 4.0 International Licence. Page numbers and pro- ceedings footer are added by the organisers. Licence details: http://creativecommons.org/licenses/by/4.0/ The main difference is that in the context of post- editing the source text is provided. A transla- tion workflow that incorporates post-editing be- gins with a source sentence, e.g. “I could still sit on the border in the very last tram to West Berlin.” and the human translator is provided with a machine-generated translation with untranslated fragments such as the previous example and some- times “fixing” the translation would simply re- quire substituting the appropriate translation for the untranslated fragment. 2 Related Tasks and Previous Approaches The L2 writing assistant task lies between the lines of machine translation and crosslingual word sense disambiguation (CLWSD) or crosslingual lexical substitution (CLS) (Lefever and Hoste, 2013; Mihalcea et al. 2010). While CLWSD systems resolve the correct semantics of the translation by providing the correct lemma in the target language, CLS at- tempts to provide also the correct form of the translation with the right morphology. Machine translation tasks focus on producing translations of whole sentences/documents while crosslingual word sense disambiguation targets a single lexical item. Previously, CLWSD systems have tried distri- butional semantics and string matching methods (Tan and Bond, 2013), unsupervised …

Content truncated to 500 words