A PROPOSED MODEL OF SELECTING FEATURES FOR CLASSIFYING ARABIC TEXT

Abstract

Classification of Arabic text plays an important role for several applications. Text classification aims at assigning predefined classes to text documents. Unstructured Arabic text can be easily processed by humans, while it is harder to be interpreted and understood by machines. So, before classifying Arabic text or documents, some pre-processing operations should be done. This work presents a proposed model for selecting features from the adopted Arabic text; i.e., documents. In this work, the words ‘text’ and ‘documents’ are used interchangeably. The adopted documents are taken from Al-Khaleej-2004 corpus. The corpus contains thousands of documents which talk about news in different domains, such as economics, as well as international, local and sport news. Some preprocessing operations are carried out to extract the highly weighted terms that best describe the content of the documents. The proposed model contains many steps to define the most relevant features. After defining the initial number of features, based on the weighted words, the steps of the model begin. The first step is based on calculating the correlation between each feature and class one. Depending on a threshold value, the most highly correlated features are chosen. This reduces the number of chosen features. The number of features is again reduced by calculating the intra-correlation between the resultant features. This is done in the second step. The third step selects the best features from among those which resulted from the second step by adopting some logical operations. The logical operations, specifically logical AND or logical OR, are applied to fuse the values of features depending on their structure, nature and semantics. The obtained features are then reduced in number. The fourth step is based on adopting the idea of document clustering; i.e., the obtained features from step three are placed in one cluster. Then, iterative operations are used to group features into two clusters. Each cluster can be further partitioned into two clusters …and so on. That partitioning is repeated till the clusters' contents are not changed. The contents of each cluster are fused together using the cosine rule. This reduces the overall number of features. This work adopts four types of classifiers; namely, Naïve Bayes (NB), Decision Tree, CART and KNN. A comparative study is carried out among the behaviors of the adopted classifiers on the selected number of features. The comparative study considers some measurable criteria; namely, precision, recall, F-measure and accuracy. This work is implemented using WEKA and MatLab software packages. From the obtained results, the best performance is achieved by using CART classifier, while the worst one is obtained by using KNN classifier

Keywords

• text classification
• text clustering
• feature selection
• arabic datasets
• machine learning methods
• performance evaluation