Effective Arabic Stemmer Based Hybrid Approach for Arabic Text Categorization

Abstract

Text pre-processing of Arabic Language is a challenge and crucial stage in Text Categorization (TC)particularly and Text Mining (TM) generally. Stemming algorithms can be employed in Arabic text pre-processing to reduces words to their stems/or root.Arabic stemming algorithms can be ranked, accordingto three category, as root-based approach (ex. Khoja); stem-based approach (ex. Larkey); and statisticalapproach (ex. N-Garm).However, no stemming of this language is perfect: The existing stemmers have asmall efficiency.In this paper, in order to improve the accuracy ofstemming and therefore the accuracy of our proposedTC system, an efficient hybrid method is proposed for stemming Arabic text. The effectiveness of theaforementioned four methods was evaluated and compared in term of the F-measure of the Naïve Bayesianclassifier and the Support Vector Machine classifier used in our TC system. The proposed stemmingalgorithm was found to supersede the other stemmingones: The obtained results illustrate that using theproposed stemmer enhances greatly the performance of Arabic Text Categorizatio

Investigate the Context Usage of Arabic Proverbs in Twitter

“…This was created by Khoja to solve the problems raised by the stemming tool (root extraction), [2], [6], [22]. According to some of previous studies, light stemming is a good stemmer for Arabic text [2], [6], [7].…”

A Survey on the Methods for Uyghur Stemming

“…Recent years, a lot of non-English mixed stemming methods are proposed, for example, a hybrid algorithm for Polish proposed by Dawid Weis [19], hybrid algorithm for Nepali stemming proposed by Chiranjibi Sitaula [20], hybrid inflectional stemmer for Gujarati proposed by Suba, Jiandani and Bhattacharyya [21], hybrid stemmer for Arabic proposed by Hadni, Ouatik and Lachkar [22 ].…”

Arabic Tweets Categorization Based on Rough Set Theory

“…Recently, Hadni et al team [7] presents an Effective Arabic Stemmer Based Hybrid Approach for Arabic Text Categorization. Note that, in any Text Categorization system the center point is the document and its representation that may impact positively or negatively on the accuracy of the system.…”

“…Figure .1 Architecture of TC System of predefined categories [6]. Several Text Categorization Systems have been conducted for yet very little researches have been done out for the Arabic Text Categorization [7]. Arabic language is a highly inflected language and it requires a processing to be manipulated, it is a Semitic language that has a very complex ared with English.…”