We propose the use of WordNet synsets
in a syntax-based reordering model for hierarchical statistical machine translation
(HPB-SMT) to enable the model to generalize to phrases not seen in the training data but that have equivalent meaning.
We detail our methodology to incorporate synsets’ knowledge in the reordering
model and evaluate the resulting WordNetenhanced SMT systems on the English-toFarsi language direction. The inclusion of
synsets leads to the best BLEU score, outperforming the baseline (standard HPBSMT) by 0.6 points absolute.
This item is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 3.0 License. View License
Funders:
Science Foundation Ireland through the CNGL Programme (Grant 12/CE/I2267) in the ADAPT Centre (www.adaptcentre.ie) at Dublin City University,, European Union Seventh Framework Programme FP7/2007-2013 under grant agreement PIAP-GA-2012-324414 (Abu-MaTran), University of Isfahan
ID Code:
23227
Deposited On:
02 May 2019 08:35 by
Thomas Murtagh
. Last Modified 02 May 2019 08:35