Paper: Comparing CRF and template-matching in phrasing tasks within a Hybrid MT system

ACL ID W14-1003
Title Comparing CRF and template-matching in phrasing tasks within a Hybrid MT system
Venue Workshop on Hybrid Approaches to Translation
Session
Year 2014
Authors

The present article focuses on improving the performance of a hybrid Machine Translation (MT) system, namely PRE- SEMT. The PRESEMT methodology is readily portable to new language pairs, and allows the creation of MT systems with minimal reliance on expensive re- sources. PRESEMT is phrase-based and uses a small parallel corpus from which to extract structural transformations from the source language (SL) to the target language (TL). On the other hand, the TL language model is extracted from large monolingual corpora. This article exam- ines the task of maximising the amount of information extracted from a very lim- ited parallel corpus. Hence, emphasis is placed on the module that learns to seg- ment into phrases arbitrary input text in SL, by extrapolating information from ...