Rule Based Method of Name Entity Recognition for Matching Allah?s Finest Names in Holy Quran

摘要

Named Entity Recognition (NER) is a standout research amongst the most imperative ones in information extraction. Furthermore, it is a sub step in many text processing applications. This study uses rule based method for matching Allah?s finest names in specific domain (Holy Qur?an), the proposed system constructs a collection of rules by regular expressions. The proposed rules match all Allah?s finest names in Holy Qur?an. Many features were considered in this study including: orthographic, N-grams and affixes. The proposed system solves the problem of appearance Allah?s finest names in Holy Qur?an as a substring in another words as well as considers the diacritics (Al-Tashkeel) in Arabic words. Similarity coefficient of names was calculated to reduce the complexity of rules base. The results showed that a rule based by regular expression is an effective, powerful and efficient technique in matching the target names with little collection of rules. This project is a valuable statistical tool that can be right applied to get statistical information about Allah?s finest names.