Stemming impact on Arabic text categorization performance: A survey

Abstract

The significant growth of online textual information has increased the demand for effective content-based Arabic text categorization methods. The categorization of Arabic texts has some challenges that need to be addressed specially when using stemming. In the literature, we found a debate among researchers about the benefits of using stemming in Arabic text categorization. Hence, we performed a study of this feature reduction method to clarify the impact of this widely used method in text mining and document classification. We also presented some Arabic text cases to deny the importance of stemming in Arabic text categorization.