kenzy444/PFE-Masters

This project is the culmination of a master's program, focusing on the development of a robust system for detecting hate speech in Arabic dialects, with a particular emphasis on the Algerian dialect. The project leverages Machine Learning (ML), Deep Learning (DL), and Large Language Models (LLMs) to achieve its objectives. # Hate Speech Detection in Arabic Dialects With Focus On Algerian Dialect This project is the culmination of a master's program, focusing on the development of a robust system for detecting hate speech in Arabic dialects, with a particular emphasis on the Algerian dialect. The project leverages Machine Learning (ML), Deep Learning (DL), and Large Language Models (LLMs) to achieve its objectives. # Project Overview The development process includes: Preprocessing Pipeline: Creating an effective preprocessing pipeline to address the unique challenges posed by the Algerian dialect. Machine Learning Baselines: Developing machine learning baselines using various text representation techniques such as TF-IDF and Word2Vec skip-gram. Deep Learning Models: Building deep learning models with different word embeddings like fastText and AraVec. Fine-Tuning LLMs: Fine-tuning seven large language models (LLMs) to improve detection accuracy. For more detailed information, please refer to the final dissertation PDF.