International Journal of Academic Information Systems Research (IJAISR)   Year: 2020 | Volume: 4 | Issue: 8 | Page No.: 6-9 Arabic Text Summarization Using AraBERT Model Using Extractive Text Summarization Approach Abdullah M. Abu Nada, Eman Alajrami, Ahemd A. Al-Saqqa, Samy S. Abu-Naser Abstract: recently, after the life of the individual changed and became more crowded with all the concerns of life, and with the diversity and the increasing of sources of knowledge on the Internet, it became difficult for us to read large texts and articles, so we are looking for the summaries of these texts before deciding dive deeply in reading. For this reason, it became urgent to provide tools to fulfill this function by extracting basic information while preserving the essence of the text. In this study, we proposed an extractive Arabic text summarizer based on a general-purpose architecture for Natural Language Generation (NLG) and Natural Language Understanding (NLU) like (AraBERT, BERT, XLNet, XLM, etc.) to summarize the Arabic document by evaluating and extracting the most important sentences at this document. Then, using the Rouge measure and human evaluation, we compared the efficiency between the proposed and other solutions to recommend what the best one we can use to summarize Arabic text and put our hands-on weak points to open the way for researchers to improve the approaches.