2 search results - page 1 / 1 » Distribution-Based Pruning of Backoff Language Models |
ACL
2000
14 years 5 days ago
2000
We propose a distribution-based pruning of n-gram backoff language models. Instead of the conventional approach of pruning n-grams that are infrequent in training data, we prune n...
EMNLP
2009
13 years 8 months ago
2009
The recent availability of large corpora for training N-gram language models has shown the utility of models of higher order than just trigrams. In this paper, we investigate meth...