Arabic Grammar Correction for Arabic Text Summaries | ||||
| Mansoura Journal for Computer and Information Sciences | ||||
| Volume 20, Issue 2, December 2025, Page 1-16 PDF (3.22 MB) | ||||
| Document Type: Original Research Articles. | ||||
| DOI: 10.21608/mjcis.2025.353920.1009 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Nada Essa   1; Mostafa Elgayar2; Eman El-daydamony1
| ||||
| 1Information Technology, Faculty of Computers and Information Sciences, Mansoura, Eygpt | ||||
| 2Department of Computer Science, Arab East Colleges, Riyadh 11583, Saudi Arabia | ||||
| Abstract | ||||
| The Arabic grammar correction is important due to the complexity of Arabic grammar, domain change, lack of training data, lack of standard databases, and many vocabularies. There is still a lot to do to get a satisfactory result in correcting the grammar for Arabic. In this paper, a new open-domain technique is presented for Arabic grammar correction. It consists of three main stages: the creation of a database for Arabic grammar representations of correct and incorrect sentences, a sequence-to-sequence gated recurrent unit encoder-decoder architecture for training the database, and testing the encoder-decoder for Arabic grammar correction. It is based on a database of correct and incorrect Arabic sentence structure using part-of-speech tags, dependency relations between words, and the features of words. In addition, the system is designed to be implemented in any domain. The Qatar Arabic Language Bank 2014 and 2015 test sets are used to test the system. The results show that the system has achieved 96.9, 94.8, and 95.83 percent for precision, recall, and F-measure. | ||||
| Keywords | ||||
| Natural Language Processing; Arabic Text Summarization; Arabic Grammar Correction | ||||
|
Statistics Article View: 78 PDF Download: 98 |
||||
