Academic Commons

Reports

Automatic Identification of Errors in Arabic Handwriting Recognition

Habash, Nizar; Roth, Ryan; Habash, Nizar Y.; Roth, Ryan M.

Arabic handwriting recognition (HR) is a challenging problem due to Arabic's connected letter forms, consonantal diacritics and rich morphology. In this paper we isolate the task of identification of erroneous words in HR from the task of producing corrections for these words. We consider a variety of linguistic (morphological and syntactic) and non-linguistic features to automatically identify these errors. We also consider a learning curve varying in two dimensions: number of segments and number of n-best hypotheses to train on. We additionally evaluate the performance on different test sets with different degrees of errors in them. Our best approach achieves a roughly ~20% absolute increase in F-score over a simple but reasonable baseline. A detailed error analysis shows that linguistic features, such as lemma models, help improve HR-error detection precisely where we expect them to: semantically inconsistent error words.

Files

More About This Work

Academic Units
Center for Computational Learning Systems
Publisher
Center for Computational Learning Systems, Columbia University
Series
CCLS Technical Report, CCLS-10-02
Published Here
December 20, 2010
Academic Commons provides global access to research and scholarship produced at Columbia University, Barnard College, Teachers College, Union Theological Seminary and Jewish Theological Seminary. Academic Commons is managed by the Columbia University Libraries.