Paper: Improvement Of A Whole Sentence Maximum Entropy Language Model Using Grammatical Features

ACL ID P01-1003
Title Improvement Of A Whole Sentence Maximum Entropy Language Model Using Grammatical Features
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2001
Authors

In this paper, we propose adding long-term grammatical information in a Whole Sentence Maximun Entropy Language Model (WSME) in order to improve the performance of the model. The grammatical information was added to the WSME model as fea- tures and were obtained from a Stochas- tic Context-Free grammar. Finally, ex- periments using a part of the Penn Tree- bank corpus were carried out and sig- nificant improvements were acheived.