Hybrid system for extracting and classifying Arabic proper names

Many applications such as information extraction systems, question answering systems, text summarization systems, information retrieval systems, etc. rely on proper names as one main tool to achieve their goals. In the Arabic language there is a big challenge for finding those proper names in the text because they do not start with capital letter as in many other languages, nor they have special sign to identify them and distinguish them from other words in the text. Little research has been conducted in this area; most efforts have been done based on a number of heuristic rules used to find names in the text, some used graphs to represent the words that might form a name and the relationships between them, some they use statistical methods for this reason. In this paper we describe a hybrid system built based on both statistical methods and predefined rules.