A suite of Arabic natural language processing tools developed by the CAMeL Lab at New York University Abu Dhabi.

UBC ARBERT and MARBERT Deep Bidirectional Transformers for Arabic

Multi-turn open-domain Arabic chatbot with a wide set of features.

The "عربي - Franko" Chrome extension is designed to provide translation services between Franko text and Arabic. It enables users to easily translate text from Franko to Arabic and vice versa.

Fine-tune BERT models to classify Arabic text by different dialects.

Egyptian / Modern Standard Arabic language identification system

Egyptian Dialect Corpus

Saudi Dialect Corpus

Nuanced Arabic Dialect Identification Shared Tasks (NADI) 2020 and 2021

Arabic dialects identification system

DiaLex - A Benchmark for Evaluating Multidialectal Arabic Word Embeddings

domain-independent multi-dialect Arabic stop words

Arabic Dialect Identification between 18 country-level Arabic dialects using QADI dataset and pretrained language model AraBERT

the backend for Qamous

using AraBert to classify different Arabic dialects. ranked fourth in WANLP2020 workshop.

We utilized a pre-trained model to classify Arabic text. After conducting extensive research, we found that MarBERT was the best model for classifying Arabic offensive tweets. It focuses on dialectal Arabic (DA) and Modern Standard Arabic (MSA). The competition involves two shared sub-tasks: detecting whether a tweet is offensive or not; and det…

A machine learning/deep learning approach to classify the dialect of arabic text.

The codebase for the "ALDi: Quantifying the Arabic Level of Dialectness of Text" paper accepted to EMNLP 2023.

A light stemmer for MDA (Moroccan Dialect Arabic) based on BPE (Byte Pair Encoding) algorithm implemented with Typescript

WIBARAB is a project in the field of Arabic dialectology. It consists of various regional sub-projects (four PhD projects) and a large database about bedouin-type dialects of Arabic. The Feature Database will be the main point of integrating the results of the sub-projects. In this repository we collect the primary data of the database in TEI/XML.