The web as a baseline: Evaluating the performance of unsupervised web-based models for a range of nlp tasks

The web as a baseline: Evaluating the performance of unsupervised web-based models for a range of nlp tasks (2004)

Cached

Download Links

[homepages.inf.ed.ac.uk]
[acl.ldc.upenn.edu]

Other Repositories/Bibliography

DBLP

by Mirella Lapata , Frank Keller

Venue:	In Proc. of Human Language Technologies - North American Chapter of the Association for Computational Linguistics (HLT-NAACL
Citations:	31 - 1 self

Versions

Version History

Metadata Version 1

Datum	Value	Source
TITLE	The web as a baseline: Evaluating the performance of unsupervised web-based models for a range of nlp tasks	INFERENCE
AUTHOR NAME	Mirella Lapata	SVM HeaderParse 0.2
AUTHOR AFFIL	Department of Computer Science; University of Sheffield	SVM HeaderParse 0.2
AUTHOR ADDR	211 Portobello St., Sheffield S1 4DP	SVM HeaderParse 0.2
AUTHOR NAME	Frank Keller	SVM HeaderParse 0.2
AUTHOR AFFIL	School of Informatics; University of Edinburgh	SVM HeaderParse 0.2
AUTHOR ADDR	2 Buccleuch Pl., Edinburgh EH8 9LW	SVM HeaderParse 0.2
ABSTRACT	Previous work demonstrated that web counts can be used to approximate bigram frequencies, and thus should be useful for a wide variety of NLP tasks. So far, only two generation tasks (candidate selection for machine translation and confusion-set disambiguation) have been tested using web-scale data sets. The present paper investigates if these results generalize to tasks covering both syntax and semantics, both generation and analysis, and a larger range of n-grams. For the majority of tasks, we find that simple, unsupervised models perform better when n-gram frequencies are obtained from the web rather than from a large corpus. However, in most cases, web-based models fail to outperform more sophisticated state-of-theart models trained on small corpora. We argue that web-based models should therefore be used as a baseline for, rather than an alternative to, standard models. 1	SVM HeaderParse 0.2
YEAR	2004	INFERENCE
VENUE	In Proc. of Human Language Technologies - North American Chapter of the Association for Computational Linguistics (HLT-NAACL	INFERENCE
VENUE TYPE	CONFERENCE	INFERENCE
PAGES	121--128	INFERENCE
CITATIONS	22 found	ParsCit 1.0