ProfessorAbraham/Amharic-Legal-RAG-Sentiment

Amharic Legal RAG + Sentiment Analysis pipeline for Ethiopian legal documents and complaints. Built for localized AI applications in Ethiopia. # 🇪🇹 Amharic Legal RAG + Sentiment Analysis **A powerful Retrieval-Augmented Generation (RAG) system for Amharic legal documents and complaints with built-in Sentiment Analysis.** This project enables intelligent Amharic language applications for Ethiopian organizations, legal aid centers, government offices, and customer support systems. ``` ``` ## ✨ Features - Full **Amharic RAG pipeline** (Retrieval-Augmented Generation) - **Sentiment Analysis** (Positive / Negative / Neutral) for Amharic text - Vector database using **Chroma** - Ready-to-use web interface (Streamlit) - Google Colab support - Clean and well-organized project structure - Focused on low-resource language support (Amharic) --- ## 📁 Project Structure ```bash ├── app/ # Streamlit web application ├── data/ # Sample legal and complaint datasets ├── src/ # Core RAG and processing logic ├── vector_db/ # Chroma vector store ├── colab_setup.ipynb # Easy Colab notebook ├── requirements.txt ├── README.md └── LICENSE ``` ## 🚀 Quick Start ### 1. Clone the repository ```bash git clone github.com cd Amharic-Legal-RAG-Sentiment ``` ### 2. Install dependencies ```bash pip install -r requirements.txt ``` ### 3. Run the application ```bash streamlit run app/main.py # or python src/main.py ``` **Google Colab Users**: Open `colab_setup.ipynb` --- ## 🛠️ Tech Stack - **Python** - **LangChain** / LlamaIndex (RAG) - **Sentence Transformers** (Amharic embeddings) - **Chroma** (Vector DB) - **Streamlit** (UI) - Hugging Face models optimized for Amharic --- ## 🎯 Use Cases - Automated analysis of customer complaints in Amharic - Legal document question-answering assistant - Sentiment monitoring for government or NGO feedback - Building localized Ethiopian AI applications --- ## 📈 Roadmap - [ ] Add mor …