--- title: "Hazem Hajj — author profile and papers" description: "Hazem Hajj, American University of Beirut: 2 papers on Elamur Research (2019–2020), h-index 27 — affiliations, topics, co-authors and who cites them in…" canonical: "https://research.elamur.ai/free/en/authors/hazem-hajj/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/hazem-hajj/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Hazem Hajj # Hazem Hajj Hazem Hajj works on natural language processing for Arabic, a language whose complex word structure, scarce tools and distance from Latin-script languages make it hard for general-purpose systems. Their papers here, from 2019 and 2020, approach this from two directions: building language models trained on Arabic alone, and creating annotated collections of dialectal social media text for opinion analysis. The work sits within [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/), with links to [Sentiment & emotion](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/) and [Multilingual & cross-lingual NLP](https://md.research.elamur.ai/free/en/topics/multilingual-nlp/). The most cited of these papers is [AraBERT](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/) (2020). It responds to the weak handling of Arabic in multilingual models by pre-training a BERT-base model on Arabic text only. Training used whole-word masking and next-sentence prediction over roughly 24GB of deduplicated news and public corpora. The earlier [ArSentD-LEV](https://md.research.elamur.ai/free/en/papers/baly-2019-arsentd-lev-multi-topic-corpus/) (2019) assembles 4,000 Levantine tweets from Jordan, Lebanon, Palestine and Syria. Each tweet is annotated for overall sentiment on a five-point scale, for the targets that opinions are aimed at, and for whether a view is expressed openly or only implied. A manual pilot study of 200 tweets shaped both the labelling scheme and the keywords used to collect the data. Taken together, the two papers move from a dialect- and topic-aware sentiment resource to a general pretrained model for Arabic language understanding. Hajj is affiliated with the [American University of Beirut](https://md.research.elamur.ai/free/en/universities/american-university-of-beirut/), where most of the papers bearing the name were written, from 2009 onward. Co-authors on the site include [Ramy Baly](https://md.research.elamur.ai/free/en/authors/ramy-baly/), who led the Levantine sentiment corpus, and [Wassim El-Hajj](https://md.research.elamur.ai/free/en/authors/wassim-el-hajj/). Hazem Hajj has 2 papers in our library, in 2019–2020, most of them on [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/), [Sentiment & emotion](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/), [Multilingual & cross-lingual NLP](https://md.research.elamur.ai/free/en/topics/multilingual-nlp/); often with [Ramy Baly](https://md.research.elamur.ai/free/en/authors/ramy-baly/), [Wassim El-Hajj](https://md.research.elamur.ai/free/en/authors/wassim-el-hajj/). 2 papers in our library · 2019–2020 current affiliation [American University of Beirut](https://md.research.elamur.ai/free/en/universities/american-university-of-beirut/) · Beirut Lebanon [OpenAlex](https://openalex.org/A5088085383) · [ORCID](https://orcid.org/0000-0002-9954-7924) · [Wikidata](https://www.wikidata.org/wiki/Q130847837) · [**h-index 27**](https://openalex.org/authors/A5088085383) · [145 works](https://openalex.org/works?filter=author.id:A5088085383) · [3,469 citations](https://openalex.org/works?filter=author.id:A5088085383&sort=cited_by_count:desc) OpenAlex, 2026-10-14 [2 works in our library](https://md.research.elamur.ai/free/en/authors/hazem-hajj/#papers) · [1,026 citations in our library](https://md.research.elamur.ai/free/en/authors/hazem-hajj/cited-by/) · 19 more works in our library · [⭐ Paper Star rank #1617](https://md.research.elamur.ai/free/en/paper-stars/page/17/#star-1617) The profile is the library's match of the name; two people with one name are not always told apart. Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Hazem+Hajj) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Hazem+Hajj) · [dblp](https://dblp.org/search?q=Hazem+Hajj) · [ACL Anthology](https://aclanthology.org/search/?q=Hazem+Hajj) ## Affiliations [🎓**American University of Beirut** Beirut · Lebanon 2009–2026 · 116 papers*current*](https://md.research.elamur.ai/free/en/universities/american-university-of-beirut/) · [🏢**Intel** Santa Clara · United States 2003–2022 · 6 papers](https://md.research.elamur.ai/free/en/companies/intel/) · [🎓**Arab International University** Ghabāghib · Syria 2021 · 4 papers](https://md.research.elamur.ai/free/en/universities/arab-international-university/) · [🎓**University of Wisconsin–Madison** Madison · United States 1996–2002 · 2 papers](https://md.research.elamur.ai/free/en/universities/university-of-wisconsin-madison/) From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 4 more institutions on single bylines are not listed. ## Topics [Semitic languages 2](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · [Sentiment & emotion 1](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/) · [Multilingual & cross-lingual NLP 1](https://md.research.elamur.ai/free/en/topics/multilingual-nlp/) ## Works by year ### 2020s [**2020** 1 paper](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/) ### 2010s [**2019** 1 paper](https://md.research.elamur.ai/free/en/papers/baly-2019-arsentd-lev-multi-topic-corpus/) ## Papers The most cited first (OpenAlex and our library). 1. **[AraBERT: Transformer-based Model for Arabic Language Understanding](https://md.research.elamur.ai/free/en/papers/antoun-2020-arabert-transformer-based-model-arabic/)** *(2020 · 1,297 citations)* Arabic has rich morphology, limited NLP resources and little in common with Latin-script languages. As a result, multilingual models such as Google's multilingual BERT represent it poorly. The authors address this by training a BERT model only on Arabic, called AraBERT. It uses the standard BERT-base architecture and… 2. **[ArSentD-LEV: A Multi-Topic Corpus for Target-based Sentiment Analysis in Arabic Levantine Tweets](https://md.research.elamur.ai/free/en/papers/baly-2019-arsentd-lev-multi-topic-corpus/)** *(2019 · 60 citations)* Most Arabic sentiment datasets assign only a single polarity label to a whole text, often on a coarse three-way scale. ## Also in our library 19 more works of theirs are in our library without a page here yet; they will appear as we publish them. ## Co-authors here 2 co-authors with a page, the most papers together first. - [Ramy Baly](https://md.research.elamur.ai/free/en/authors/ramy-baly/) - [Wassim El-Hajj](https://md.research.elamur.ai/free/en/authors/wassim-el-hajj/) ## Cited by Papers of our library cite their works 1,026 times (of the 140,884 papers whose bibliographies we read); [196 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/hazem-hajj/cited-by/).