A Comprehensive Review of Arabic Text Summarization

The explosion of online and offline data has changed how we gather, evaluate, understand data. It is frequently difficult time-consuming to comprehend large text documents extract crucial information from them. Text summarization techniques address the mentioned problems by compressing long texts while retaining their essential contents. These rely on fast delivery filtered, high-quality content users. Due massive amounts generated technology various sources, automated large-scale challenging. There are three types automatic techniques: extractive, abstractive, hybrid. Regardless these previous techniques, summaries a way produced human experts. Although Arabic widely spoken language that used for sharing web, limited still immature because several problems, including language’s morphological structure, variety dialects, lack adequate sources. This paper reviews approaches recent deep learning models this approach. Additionally, it focuses existing datasets approaches, which also reviewed, along with characteristics limitations. most often metrics quality evaluation ROUGE1, ROUGE2, ROUGE L, Bleu. challenges encountered during summarizing methods solutions proposed in each approach analyzed. Many have such as golden tokens testing, being out vocabulary (OOV) words, repeating summary sentences, standard systematic methodologies architectures, complexity language. Finally, providing required corpora, improving using semantic representations, rouge abstractive summarization, adopt them studies an demand.