شاعر بلا مشاعر: دراسة تحليلية في الشعر العربي الآلي بين لغة الرسول ولغة الإعلام الاجتماعي

A Poet Without Emotions: An Analytical Study of Automated Arabic Poetry Between the Language of the Prophet and Social Media Language

Abstract

This research explores Natural Language Processing (NLP) as a pivotal domain within Artificial Intelligence (AI), with a particular focus on the application of Deep Learning techniques in Arabic text generation. The study investigates the evolution of Arabic language expression, from the eloquence of Prophetic traditions to the rapid, informal nature of social media discourse. As a case study, we attempt to generate text that emulates the poetic style of Nizar Qabbani, examining linguistic transitions between classical Arabic and contemporary digital expressions.We provide a historical overview of NLP, outlining its origins and major developmental milestones. Key applications such as machine translation, sentiment analysis, opinion mining, and automated question-answering systems are explored alongside core NLP subfields, including text classification, tokenization, syntactic parsing, semantic understanding, and text generation. Special emphasis is placed on the linguistic distinctions between Quranic Arabic, Hadith literature, and modern digital communication. A significant portion of this study is dedicated to the integration of Deep Learning into NLP, particularly the use of Recurrent Neural Networks (RNNs) in generating Arabic poetry. The research evaluates the interplay between linguistic purity and computational text generation, comparing rhetorical elements of Prophetic speech with the concise and informal nature of social media language. Furthermore, open-source tools and methodologies for Arabic language processing are discussed, providing a comparative analysis of classical and digital text processing techniques. By referencing key resources, this study offers valuable insights for researchers interested in Arabic NLP and its broader implications.