A Pipeline Approach to Supervised Error Correction for the QALB-2014 Shared Task

Nadi Tomeh, Nizar Habash, Ramy Eskander, Joseph Le Roux

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper describes our submission to the ANLP-2014 shared task on automatic Arabic error correction. We present a pipeline approach integrating an error detection model, a combination of character- and word-level translation models, a reranking model and a punctuation insertion model. We achieve an F1 score of 62.8% on the development set of the QALB corpus, and 58.6% on the official test set.

Original languageEnglish (US)
Title of host publicationANLP 2014 - EMNLP 2014 Workshop on Arabic Natural Language Processing, Proceedings
EditorsNizar Habash, Stephan Vogel
PublisherAssociation for Computational Linguistics (ACL)
Pages114-120
Number of pages7
ISBN (Electronic)9781937284961
StatePublished - 2014
EventEMNLP 2014 Workshop on Arabic Natural Language Processing, ANLP 2014 - Doha, Qatar
Duration: Oct 25 2014 → …

Publication series

NameANLP 2014 - EMNLP 2014 Workshop on Arabic Natural Language Processing, Proceedings

Conference

ConferenceEMNLP 2014 Workshop on Arabic Natural Language Processing, ANLP 2014
Country/TerritoryQatar
CityDoha
Period10/25/14 → …

ASJC Scopus subject areas

  • Language and Linguistics
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Software

Fingerprint

Dive into the research topics of 'A Pipeline Approach to Supervised Error Correction for the QALB-2014 Shared Task'. Together they form a unique fingerprint.

Cite this