A Rule-based Named Entity Extraction Method and Syntactico-Semantic Annotation for Arabic Language

摘要

There is a widely held belief in the natural language processing (NLP) and computational linguistics communities that knowledge recognition such us Named Entities (NE) recognition is a significant step toward improving important applications, e.g., question answering and natural language understanding (NLU). In this paper, we present an NE recognition system for Modern Standard Arabic using the NooJ platform. This system exploits many aspects of the rich morphological features of the language. The experiments on the pilot Arabic Propbank data show that our system based on linguistic rules produces a global NE recognition F-measure score of 87%, which improves the current state of the art in Arabic NE recognition.