ArCAR: A Novel Deep Learning Computer-Aided Recognition for Character-Level Arabic Text Representation and Recognition

Arabic text classification is a process to simultaneously categorize the different contextual Arabic contents into a proper category. In this paper, a novel deep learning Arabic text computer-aided recognition (ArCAR) is proposed to represent and recognize Arabic text at the character level. The input Arabic text is quantized in the form of 1D vectors for each Arabic character to represent a 2D array for the ArCAR system. The ArCAR system is validated over 5-fold cross-validation tests for two applications: Arabic text document classification and Arabic sentiment analysis. For document classification, the ArCAR system achieves the best performance using the Alarabiya-balance dataset in terms of overall accuracy, recall, precision, and F1-score by 97.76%, 94.08%, 94.16%, and 94.09%, respectively. Meanwhile, the ArCAR performs well for Arabic sentiment analysis, achieving the best performance using the hotel Arabic reviews dataset (HARD) balance dataset in terms of overall accuracy and F1-score by 93.58% and 93.23%, respectively. The proposed ArCAR seems to provide a practical solution for accurate Arabic text representation, understanding, and classification.

The effects of Pre-Processing Techniques on Arabic Text Classification

In the last two decades, the amount of available Arabic text data on the World Wide Web is dramatically growing, making it the fourth most used language on the web. Accordingly, the demand for efficient Arabic text classification is increasing, especially for web page content filtering, information retrieval, and e-mail spam detection. Several Machine Learning algorithms have been implemented to classify Arabic documents. However, the results achieved are not comparable with those obtained in other languages such as English, primarily when using preprocessing techniques that do not take into consideration the Arabic language features. This paper investigates the impact of wisely selected preprocessing techniques on the efficiency of different text classification algorithms. The effects of stop words removal, stemming, lemmatization, and all possible combinations are examined. The reported results (+10.75% to +28.73%) prove the effectiveness of using these techniques either individually or in combination.