期刊名称：International Journal of Advances in Soft Computing and Its Applications
印刷版ISSN：2074-8523
出版年度：2020
卷号：12
期号：2
页码：1-15
出版社：International Center for Scientific Research and Studies
摘要：Authorship attribution (AA) helps to identify the original author of unseen text by extracting and using several features that can be used to differentiate the writing style of one author from others. Many classification methods have been applied for identifying the authorship attribution for different languages. Few works have been done for Arabic authorship attribution. This paper aims to investigate the performance of deep learning-based artificial neural network for identifying the attribution of authors for Arabic text. A dataset that includes 4686 Arabic texts for 15 different authors was used in this study. The performance of deep learning method was compared with several machine learning methods. The experimental results showed the superior performance of deep learning for authorship attribution in Arabic language using F-score, accuracy, precision and recall measures.
