AN IMPROVED ARABIC WORD�S ROOTS EXTRACTION METHOD USING N-GRAM TECHNIQUE | Science Publications

Journal of Computer Science

Ashraf Odeh¹  Aymen Abu-Errub¹  Hayel Khafajeh¹  Nidal Yousef¹ 

关键词: Arabic Root Extraction;    Natural Language Processing;    N-Gram;   

DOI  :  10.3844/jcssp.2014.716.719

学科分类：计算机科学（综合）

来源: Science Publications

【 摘 要 】

Arabic language is distinguished by its morphological richness, which forces the workers in the field of Arabic language Processing (i.e., information retrieval, document’s classification, text summarizing) to deal with many words that seem to be different but in reality they came from an identical root word. One of the methods to overcome this problem is to return the words to their roots. This research aims to provide a new algorithm, that returns roots of Arabic words using n-gram technique without using morphological rules in order to avoid the complexity arising from the morphological richness of the language in one hand and the multiplicity of morphological rules in other hand. The proposed algorithm uses a list that contains over 4,500 identical roots words.

【 授权许可】

Unknown