Arabic Text Categorization

Abstract

In this paper, the researcher compared the performance of two classifiers for Arabic text classification. Naïve Bayes and Key Nearest Neighbor (KNN) were used to classify the documents. These documents which were not classified were preprocessed by removing stop words and punctuation marks from them. The word in each document was presented as a vector . These vectors were used in WEKA tool to give the results. The accuracy of two algorithms was compared using precision, recall, f-measure. The results showed that the accuracy Naïve Bayes algorithm was better than Key Nearest Neighbor( KNN) algorithm .