J. Botha, C. Dyer, and P. Blunsom. Proceedings of the 24th International Conference on Computational Linguistics (COLING-12), Mumbai, India, (2012)
Abstract
In this work we address the challenge of augmenting n-gram language models according to prior linguistic intuitions. We argue that the family of hierarchical Pitman-Yor language models is an attractive vehicle through which to address the problem, and demonstrate the approach by proposing a model for German compounds. In our empirical evaluation the model outperforms a modified Kneser-Ney n-gram model in test set perplexity. When used as part of a translation system, the proposed language model matches the baseline BLEU score for English→German while improving the precision with which compounds are output. We find that an approximate inference technique inspired by the Bayesian interpretation of Kneser-Ney smoothing (Teh, 2006) offers a way to drastically reduce model training time with negligible impact on translation quality.
%0 Conference Paper
%1 botha2012bayesian
%A Botha, Jan A.
%A Dyer, Chris
%A Blunsom, Phil
%B Proceedings of the 24th International Conference on Computational Linguistics (COLING-12)
%C Mumbai, India
%D 2012
%K 2012 bayesian coling compounds
%T Bayesian Language Modelling of German Compounds
%U http://www1.ccls.columbia.edu/~habash/coling-2012-citations/PAPERS022%202.pdf
%X In this work we address the challenge of augmenting n-gram language models according to prior linguistic intuitions. We argue that the family of hierarchical Pitman-Yor language models is an attractive vehicle through which to address the problem, and demonstrate the approach by proposing a model for German compounds. In our empirical evaluation the model outperforms a modified Kneser-Ney n-gram model in test set perplexity. When used as part of a translation system, the proposed language model matches the baseline BLEU score for English→German while improving the precision with which compounds are output. We find that an approximate inference technique inspired by the Bayesian interpretation of Kneser-Ney smoothing (Teh, 2006) offers a way to drastically reduce model training time with negligible impact on translation quality.
@inproceedings{botha2012bayesian,
abstract = {In this work we address the challenge of augmenting n-gram language models according to prior linguistic intuitions. We argue that the family of hierarchical Pitman-Yor language models is an attractive vehicle through which to address the problem, and demonstrate the approach by proposing a model for German compounds. In our empirical evaluation the model outperforms a modified Kneser-Ney n-gram model in test set perplexity. When used as part of a translation system, the proposed language model matches the baseline BLEU score for English→German while improving the precision with which compounds are output. We find that an approximate inference technique inspired by the Bayesian interpretation of Kneser-Ney smoothing (Teh, 2006) offers a way to drastically reduce model training time with negligible impact on translation quality.},
added-at = {2013-02-02T02:33:38.000+0100},
address = {Mumbai, India},
author = {Botha, Jan A. and Dyer, Chris and Blunsom, Phil},
biburl = {https://www.bibsonomy.org/bibtex/2a21270743b78df32727df0411c297f43/seandalai},
booktitle = {Proceedings of the 24th International Conference on Computational Linguistics (COLING-12)},
interhash = {303dd0a9779206ebb2da57cc25f91856},
intrahash = {a21270743b78df32727df0411c297f43},
keywords = {2012 bayesian coling compounds},
timestamp = {2013-02-02T02:33:38.000+0100},
title = {Bayesian Language Modelling of German Compounds},
url = {http://www1.ccls.columbia.edu/~habash/coling-2012-citations/PAPERS022%202.pdf},
year = 2012
}