Determining Case in Arabic: Learning Complex Linguistic Behavior Requires Complex Linguistic Features

This paper discusses automatic determination of case in Arabic. This task is an important part and major source of errors in full diacritization of Arabic. We use a goldstandard syntactic tree, and obtain an error rate of about 4.2%, with a machine learning based system outperforming a system using hand-written rules. A careful error analysis suggests that when we account for annotation errors in the gold standard, the error rate drops to 0.9%, with the hand-written rules outperforming the machine learningbased system.

• Type: Conference
• Year: 2007
• Where: EMNLP
• Authors: Nizar Habash, Ryan Gabbard, Owen Rambow, Seth Kulick, Mitchell P. Marcus
• Keywords: Careful Error Analysis, EMNLP 2007, Error Rate, Hand-written Rules, Natural Language Processing