The Arabic Citation Index (ARCI) was launched in 2020. This article provides an overview of the
scientific literature contained in this new database and explores its possible usage in research evaluation.
I used the metadata available in scientific publications to characterise ARCI's coverage. First, I describe
the data and the methodologies that were used to conduct this study. As of May 2022, ARCI had indexed
138,283 scientific publications published between 2015 and 2020. Second, I investigate the distributions
of the indexed literature at various levels (research domains, countries, languages, open access). Articles
make up nearly all of the documents indexed with a share of 99% of the ARCI database. The Arts &
Humanities and Social Sciences fields have the highest concentration of publications. Most indexed
journals are published in Egypt, Algeria, Iraq, Jordan, and Saudi Arabia. About 8% of publications in
ARCI are published in languages other than Arabic. Then, I use an unsupervised machine learning
model, LDA (Latent Dirichlet Allocation), and the text mining algorithm of VOSviewer to uncover the
main topics in ARCI. These methods provide a better understanding of ARCI's thematic structure. Next,
I discuss how ARCI can serve as a complement to global standards in the context of a more inclusive
research evaluation. Finally, I suggest few research opportunities after discussing the findings of this
study.