Pre-trained Transformers for Arabic Language Understanding and Generation (Arabic BERT, Arabic GPT2, Arabic ELECTRA)

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

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

Arabic_Dialect_Identification_NLP-AIM-Task

Simple Script to undo Farasa Segmentation, compatible with AraBERT pre-segmentation

Easy to use extractive text summarization with AraBERT

Mental health diagnosis tool using NLP and ML for Arabic inputs, with a Laravel web application interface

After collecting 40 thousand tweets and preprocessing it, I used word embeddings with arabert and tf-idf along with two neural network architectures and 5 machine learning algorithms. Due to the huge size of the dataset, I chose Amazon SageMaker to train the models

Arabic Dialect Sentimenal Analysis

This is an experiment for Qur'an QA for the shared task at the OSCAT workshop

Diacritics are short vowels with a constant length that are spoken. The same word in the Arabic language can have different meanings and different pronunciations based on how it is diacritized. In this project, we implement a pipeline to predict the diacritic of each character in an Arabic text using Natural Language Processing techniques.

Emotion Prediction in Arabic Text

Sentiment analysis with arabert in Tunisian dialect

An advanced Arabic fake news detection model using LSTM and AraBERT. This project leverages the Arabic Fake News Dataset (AFND) to classify news articles as credible, not credible, or undecided. Includes preprocessing steps, model building, and evaluation using TensorFlow.

Fine-tuning / pre-training AraElectra on a specific domain for QA system.

Disambiguation Study for Arabic Applied on Text Classification

Many countries speak Arabic; however, each country has its own dialect, the aim of this project is to build a model that predicts the dialect given the text.

Dialectical Arabic Sentiment Analysis

Used “aubmindlab/bert-base-arabertv2” from Aub-mind AraBERT to create a simple Arabic text tokenizer.