An unsupervised method for detecting grammatical errors

  • Authors:
  • Martin Chodorow;Claudia Leacock

  • Affiliations:
  • Hunter College of CUNY, New York, NY;Educational Testing Service, Princeton, NJ

  • Venue:
  • NAACL 2000 Proceedings of the 1st North American chapter of the Association for Computational Linguistics conference
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present an unsupervised method for detecting grammatical errors by inferring negative evidence from edited textual corpora. The system was developed and tested using essay-length responses to prompts on the Test of English as a Foreign Language (TOEFL). The errorrecognition system, ALEK, performs with about 80% precision and 20% recall.