Towards a new Approach for Arabic root extraction: Exploit relations between the word letters and their placement in the word for Arabic root extraction

This paper presents a new root-extraction approach for Arabic words. The approach tries to assign for Arabic words a unique root without relying on a database of word roots, a list of word patterns or a list of all the prefixes and the suffixes of the Arabic words. Unlike most of Arabic rule-based stemmers, it tries to predict the root-letters positions one by one based on some rules and relations among the word letters and their placement in the word. This paper focuses on two parts of the approach. The first one introduces some rules to distinguish between the Arabic definite article (È@āl ) and the permanent component (È@āl ) that may found in any Arabic word. The second one classifies Arabic letters in to groups according to their positions in the word. The proposed approach is a system composed of several modules used to extract the word root. The approach has been evaluated using the Holy Quran words. The evaluation results show a promising root extraction algorithm.