Improving Trigram Language Modeling with The World Wide Web

Improving Trigram Language Modeling with The World Wide Web (2001)

by Xiaojin Zhu , Ronald Rosenfeld

Venue:	Acoustics, Speech, and Signal Processing, 2001. Proceedings.(ICASSP’01
Citations:	28 - 0 self

User correction supplied by SystemCorrections

Datum	Value	Source
TITLE	Improving Trigram Language Modeling with The World Wide Web	user correction - Legacy Corrections
AUTHOR NAME	Xiaojin Zhu	SVM HeaderParse 0.1
AUTHOR AFFIL	School of Computer Science, Carnegie Mellon University	SVM HeaderParse 0.2
AUTHOR ADDR	5000 Forbes Avenue, Pittsburgh, PA 15213, USA	SVM HeaderParse 0.1
AUTHOR NAME	Ronald Rosenfeld	SVM HeaderParse 0.1
AUTHOR AFFIL	School of Computer Science, Carnegie Mellon University	SVM HeaderParse 0.2
AUTHOR ADDR	5000 Forbes Avenue, Pittsburgh, PA 15213, USA	SVM HeaderParse 0.1
ABSTRACT	We propose a novel method for using the World Wide Web to acquire trigram estimates for statistical language modeling. We submit an N-gram as a phrase query to web search engines. The search engines return the number of web pages containing the phrase, from which the N-gram count is estimated. The N-gram counts are then used to form web-based trigram probability estimates. We discuss the properties of such estimates, and methods to interpolate them with traditional corpus based trigram estimates. We show that the interpolated models improve speech recognition word error rate significantly over a small test set. 1.	user correction - Legacy Corrections
YEAR	2001	INFERENCE
VENUE	Acoustics, Speech, and Signal Processing, 2001. Proceedings.(ICASSP’01	INFERENCE
VENUE TYPE	CONFERENCE	INFERENCE
PAGES	533--536	INFERENCE
CITATIONS	13 found	ParsCit 1.0