Dia AbuZeina

Stemming impact on Arabic text categorization performance: A survey

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.

Beyond vector space model for hierarchical Arabic text classification: A Markov chain approach

Abstract The vector space model (VSM) is a textual representation method that is widely used in documents classification. However, it remains to be a space-challenging problem. One attempt to alleviate the space problem is by using dimensionality reduction techniques, however, such techniques have deficiencies such as losing some important information. In this paper, we propose a novel text classification method that neither uses VSM nor dimensionality reduction techniques. The proposed method is a space efficient method that utilizes the first order Markov model for hierarchical Arabic text classification. For each category and sub-category, a Markov chain model is prepared based on the neighboring characters sequences. The prepared models are then used for scoring documents for classification purposes. For evaluation, we used a hierarchical Arabic text data collection that contains 11,191 documents that belong to eight topics distributed into 3-levels. The experimental results show that the Markov chains based method significantly outperforms the baseline system that employs the latent semantic indexing (LSI) method. That is, the proposed method enhances the F1-measure by 3.47%. The novelty of this work lies on the idea of decomposing words into sequences of characters, which found to be a promising approach in terms of space and accuracy. Based on our best knowledge, this is the first attempt to conduct research for hierarchical Arabic text classification with such relatively large data collection.

Employing fisher discriminant analysis for Arabic text classification

Abstract Fishers discriminant analysis; also called linear discriminant analysis (LDA), is a popular dimensionality reduction technique that is widely used for features extraction. LDA aims at finding an optimal linear transformation based on maximizing a class separability. Even though LDA shows useful results in various pattern recognition problems, such as face recognition, less attention has been devoted to employing this technique in Arabic information retrieval tasks. In particular, the sizable feature vectors in textual data enforces to implement dimensionality reduction techniques such as LDA. In this paper, we empirically investigated an LDA based method for Arabic text classification. We used a corpus that contains 2,000 documents belonging to five categories. The experimental results showed that the performance of semantic loss LDA based method was almost the same as the semantic rich singular value decomposition (SVD), and that is indication that LDA is a promising method for text mining applications.