marbert

UBC ARBERT and MARBERT Deep Bidirectional Transformers for Arabic

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…

Dialectal Arabic Aspect-Based Sentiment Analysis using MARBERT

This Jupyter Notebook demonstrates Arabic text classification using the MARBERT model, incorporating cross-validation to ensure robust performance evaluation.

Ai Application For Enhancing Customer Services

Arabic dialect classification using TF-IDF + Logistic Regression and MARBERT, with comparative evaluation and an interactive Gradio interface.

FineTuning LLMs and MLs models

the following notebook is to train MARBERT Model for Arabic tweet classification

Production Arabic complaint triage — 3 MARBERT models classify, a rule engine decides the action, an LLM only explains. FastAPI · Docker · Railway.

BAREC-ST-2026. The Second Shared Task on Sentence-level Readability Assessment (Open Track): An NLP competition on fine-grained Arabic readability classification across 19 levels.

🎯 Masha'erohom: State-of-the-art Arabic sentiment analysis achieving 90% accuracy using ensemble deep learning (Bi-LSTM + Bi-GRU + MARBERTv2 + Random Forest). Published research with web application for real-time emotion detection.

Arabic stance detection system for StanceEval-2026 on Mawqif-v2, combining fine-tuned Arabic encoders, retrieval-augmented few-shot LLMs, LoRA adapters, and a per-class stacking ensemble for seen and unseen target evaluation.

BAREC-ST-2026. The Second Shared Task on Sentence-level Readability Assessment (Strict Track): An NLP competition on fine-grained Arabic readability classification across 19 levels.

Arabic sentiment analysis model that classifies text into different categories using MARBERT(Transformers)

We compare the performance of multiple BERT-based models for the task of Emotion recognition in Arabic Tweets.