Qabas: An Open-Source Arabic Lexicographic Database
This provides a foundational resource for Arabic NLP researchers and developers, addressing a domain-specific bottleneck in language processing.
The authors tackled the lack of a comprehensive Arabic lexicon for NLP by creating Qabas, an open-source database that synthesizes 110 lexicons and links to 12 corpora, resulting in about 58K lemmas, making it the most extensive Arabic lexicon available.
We present Qabas, a novel open-source Arabic lexicon designed for NLP applications. The novelty of Qabas lies in its synthesis of 110 lexicons. Specifically, Qabas lexical entries (lemmas) are assembled by linking lemmas from 110 lexicons. Furthermore, Qabas lemmas are also linked to 12 morphologically annotated corpora (about 2M tokens), making it the first Arabic lexicon to be linked to lexicons and corpora. Qabas was developed semi-automatically, utilizing a mapping framework and a web-based tool. Compared with other lexicons, Qabas stands as the most extensive Arabic lexicon, encompassing about 58K lemmas (45K nominal lemmas, 12.5K verbal lemmas, and 473 functional-word lemmas). Qabas is open-source and accessible online at https://sina.birzeit.edu/qabas.