Arabic Sentiment Analysis Using AraBERT, LSTM & Machine Learning

## Project Overview

Developed an Arabic Sentiment Analysis project using Natural Language Processing (NLP) and machine learning techniques to classify Arabic text as positive or negative.

## Technologies Used

* Python

* Natural Language Processing (NLP)

* AraBERTv2 (Transformer Model)

* LSTM (Deep Learning)

* TF-IDF and Logistic Regression

* Hugging Face Transformers

* TensorFlow and Scikit-learn

## Key Features

* Arabic text preprocessing and cleaning.

* Preparation and splitting of labeled text datasets.

* Training and evaluation of multiple sentiment classification models.

* Performance comparison using accuracy, precision, recall, and F1-score.

* Sentiment prediction for new Arabic text using a fine-tuned AraBERTv2 model.

## Project Deliverables

The project includes a Jupyter Notebook containing the implementation and a PowerPoint presentation explaining the methodology, models, evaluation, results, and future improvements.

This project demonstrates practical experience in NLP, deep learning, transformer-based models, and Arabic text classification.