A Multi-Agent System for Detecting and Correcting “Hidden” Spelling Errors in Arabic Texts

Chiraz Ben Othmane Zribi, Fériel Ben Fraj, Mohamed Ben Ahmed

Abstract

In this paper, we address the problem of detecting and correcting hidden spelling errors in Arabic texts. Hidden spelling errors are morphologically valid words and therefore they cannot be detected or corrected by conventional spell checking programs. In the work presented here, we investigate this kind of errors as they relate to the Arabic language. We start by proposing a classification of these errors in two main categories: syntactic and semantic, then we present our multi-agent system for hidden spelling errors detection and correction. The multi-agent architecture is justified by the need for collaboration, parallelism and competition, in addition to the need for information exchange between the different analysis phases. Finally, we describe the testing framework used to evaluate the system implemented.

References

  1. Ben Othmane Zribi C. De la synthèse lexicographique à la détection et à la correction des graphies fautives arabes. Thèse de doctorat, Université de Paris XI, Orsay, 1998.
  2. Ben Othmane Zribi C. and Ben Ahmed M. Le contexte au service de la correction des graphies fautives arabes. TALN'03, Nantes, 11-13 Juin 2003.
  3. Bigert J. and Knutsson O. Robust Error Detection: A Hybrid Approach Combining Unsupervised Error Detection and Linguistic Knowledge. In Proceedings of Robust Methods in Analysis of Natural Language Data (ROMAND'02), Frascati, Italie, 2002.
  4. Golding A. R. A bayesian hybrid method for context- sensitive spelling correction. In Proceedings of the Third Workshop on Very Large Corpora, Cambridge, Massachusetts, USA, pages 39-53, 1995.
  5. Golding A. R. et Dan Roth. Applying winnow to context-sensitive spelling correction. In Lorenza Saitta (ed.) Machine Learning: Proceedings of the 13th International Conference. Bari, Italie, pp. 182-190, 1996.
  6. Golding A. R. et Dan Roth. A winnow-based approach to context-sensitive spelling correction. Machine Learning, 34(1-3), 107-130, 1999.
  7. Verberne S. Context sensitive spell checking based on word trigram probavilities. Mémoire de Mastère, Université de Nijmegen, 2002.
  8. Xiaolong W., Jianhua L. Combine trigram and automatic weight distribution in Chinese spelling error correction. Journal of computer Science and Technology, Volume 17 Issue 6, Province, China, 2001.
Download


Paper Citation


in Harvard Style

Ben Othmane Zribi C., Ben Fraj F. and Ben Ahmed M. (2005). A Multi-Agent System for Detecting and Correcting “Hidden” Spelling Errors in Arabic Texts . In Proceedings of the 2nd International Workshop on Natural Language Understanding and Cognitive Science - Volume 1: NLUCS, (ICEIS 2005) ISBN 972-8865-23-6X, pages 149-154. DOI: 10.5220/0002556601490154


in Bibtex Style

@conference{nlucs05,
author={Chiraz Ben Othmane Zribi and Fériel Ben Fraj and Mohamed Ben Ahmed},
title={A Multi-Agent System for Detecting and Correcting “Hidden” Spelling Errors in Arabic Texts},
booktitle={Proceedings of the 2nd International Workshop on Natural Language Understanding and Cognitive Science - Volume 1: NLUCS, (ICEIS 2005)},
year={2005},
pages={149-154},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002556601490154},
isbn={972-8865-23-6X},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 2nd International Workshop on Natural Language Understanding and Cognitive Science - Volume 1: NLUCS, (ICEIS 2005)
TI - A Multi-Agent System for Detecting and Correcting “Hidden” Spelling Errors in Arabic Texts
SN - 972-8865-23-6X
AU - Ben Othmane Zribi C.
AU - Ben Fraj F.
AU - Ben Ahmed M.
PY - 2005
SP - 149
EP - 154
DO - 10.5220/0002556601490154