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

ACL ID W14-3614
Title A Pipeline Approach to Supervised Error Correction for the QALB-2014 Shared Task
Venue Workshop on Arabic Natural Language Processing
Session
Year 2014
Authors

This paper describes our submission to the ANLP-2014 shared task on auto- matic Arabic error correction. We present a pipeline approach integrating an er- ror detection model, a combination of character- and word-level translation mod- els, a reranking model and a punctuation insertion model. We achieve an F 1 score of 62.8% on the development set of the QALB corpus, and 58.6% on the official test set.