Mohamed Eldesouki | Hamad Bin Khalifa University
Hamad Bin Khalifa University


Mohamed Ibrahim Eldesouki Tawfik Mohamed

Mohamed Ibrahim Eldesouki Tawfik Mohamed

Research Associate
Qatar Computing Research Institute QCRI
Arabic Language Technologies

  • Phone44547781
  • Office locationResearch Complex, QCRI, B1, ALT section, Open Space


Mohamed Eldesouki is a research associate and a software engineer with senior Researching and Developing skills and a wide technical and academic background in the fields of Arabic Language Technologies (ALT) and Machine Learning (ML) with a broad experience covering a diverse set of software development tools, languages and methodologies, gathered throughout almost 11 years of experience.

Mohamed has participated in designing, implementing and maintaining of NLP-related project such as Search Engines, Content-based Recommendation Systems, Auto-complete and correction for mobile devices, and Intelligent Tutoring Systems (ITS) for both Arabic and English content .

With a strong academic foundation in Information Retrieval (IR) and Information Extraction (IE) gained through the conducted research activities and studies of the MASTER’S DEGREE and deep linguistic analysis, Natural language Understanding NLU and Machine comprehension of text currently through his work at Qatar Computing Research Institute (QCRI). 


Research Interests

  • Natural Language Processing
  • Machine Learning
  • Information Retrieval
  • Machine Com-prehension of Text
  • Data Discovery and Analytics


Research Associate

Qatar Computing Research Institute (QCRI), Doha, Qatar

DEc 2015 - Present
  • Senior Research Software Engineer

    OMS Company, Cairo, Egypt

    Jan 2015 - Dec 2015
  • Senior Research Software Engineer

    Taya IT Company, Cairo, Egypt

    Aug 2013 - Dec 2014
  • Research Engineer

    MGD Company, Cairo, Egypt

    Jul 2012 - Aug 2013
  • Research Assistant

    Institute of Statistical Studies and Research (ISSR), Cairo University, Cairo, Egypt

    Jul 2009 - Jun 2012


Master (M.Sc) in Computer Science

Cairo University; Giza, Egypt

  • Bachelor of Computers and Information

    Cairo University; Giza, Egypt


Selected Publications

  • Salvatore Romeo, Giovanni Da San Martino, Yonatan Belinkov, Alberto Barŕon-Cedẽno, Mohamed Eldesouki, Kareem Darwish, Hamdy Mubarak, James Glass, and Alessandro Moschitti

    Language Processing and Learning Models for Community Question Answering in Arabic, Information Processing & Management, Volume 56, Issue 2, March2019, Pages 274-290. [ARTICLE]

  • Mohamed Eldesouki, Suwon Shon, and Ahmed Ali, James Glass

    QCRI-MITLive Arabic Dialect Identification System, QCRI - MIT CSAIL Annual Meeting Qatar National Convention Centre March 27-28, 2018. [POSTER]

  • Mohamed Eldesouki, Suwon Shon, and Ahmed Ali

    QCRI-MIT Live Arabic Dialect Identification System, ICASSP, Calgary, Canada. [DEMO]

  • Kareem Darwish, Hamdy Mubarak,Mohamed Eldesouki, Ahmed Abdelali, Younes Samih, Randah Alharbi, Mohammed Attia, Walid Magdy, and Laura Kallmeyer

    Multi-Dialect Arabic POS Tagging: A CRF Approach, In 11th edition of the Language Resources and Evaluation Conference (LREC), 7-12 May 2018, Miyazaki (Japan).

  • Younes Samih, Mohamed Eldesouki, Mohammed Attia,Kareem Darwish, Ahmed Abdelali, Hamdy Mubarak and Laura Kallmeyer

    Learning from Relatives: Unified Dialectal Arabic Segmentation. In Proceedings of the 21st Conference on Computational Natural Language Learning (CoNLL 2017), Vancouver, Canada, 432-441.

  • Walid Magdy, andMohamed Eldesouky

    ClassStrength: A Multilingual Tool for Tweets Classification, Proc. of the 2017 IEEE/ACM International Conference on Advances in social networks analysis and mining (ASONAM), Sydney, Australia, 593-596.

  • Mohamed Eldesouki, Younes Samih, Ahmed Abdelali, Mohammed Attia, Hamdy Mubarak, Kareem Darwish and Kallmeyer Laura

    Arabic Multi-Dialect Segmen-tation: bi-LSTM-CRF vs. SVM,

  • Younes Samih, Mohammed Attia, Mohamed Eldesouki, Ahmed Abdelali, Hamdy Mubarak, Laura Kallmeyer and Kareem Darwish

    A Neural Architecture for Dialectal Arabic Segmentation. In Proc. of The 3rd Arabic Natural Language Processing Workshop (WANLP-2017) co-located with EACL 2017, Valencia, Spain, pages 46-54.

  • Kareem Darwish, Hamdy Mubarak, Ahmed Abdelali and Mohamed Eldesouki

    Arabic POS Tagging: Don’t Abandon Feature Engineering Just Yet, In Proc. of The 3rdArabic NLP Workshop (WANLP-2017) co-located with EACL 2017, Valencia, Spain, P.130.

  • Mohamed Eldesouki, Fahim Dalvi, Hassan Sajjad, and Kareem Darwish

    QCRI@ DSL 2016: Spoken Arabic Dialect Identification Using Textual Features, Proc. of the3rd Workshop on NLP for Similar Languages, Varieties and Dialects, (VarDial 3), Osaka,Japan, P. 221.

  • Mohamed I. Eldesouki,An Intelligent Agent for Arabic Web Information Retrieval

    Master’s Thesis, Cairo University.

  • Mohamed Eldesouki, Waleed Arafa, Kareem Darwish, Mervat H. Gheith

    Rep-resenting Arabic Documents Using Controlled Vocabulary Extracted from Wikipedia, InProc. of The 11th Conference on Language Engineering (ESOLEC’11), Cairo, Egypt.

  • Mohamed Eldesouki, Waleed Arafa, Kareem Darwish, Mervat Gheith

    UsingWikipedia for Retrieving Arabic Documents. In Proceedings of Arabic Language Tech-nology International Conference (ALTIC 2011), Alexandria, Egypt.

  • Mohamed I. Eldesouki, Waleed M. Arafa, Kareem M. Darwish

    Stemming tech-niques of Arabic Language: Comparative Study from the Information Retrieval Perspec-tive, The Egyptian Computer Journal. 36(1):30-49.