YonatanAwoke/Ethiopian_History_RAG

A Python tool to collect, clean, and structure Ethiopian history content from Wikipedia. Outputs JSON with rich metadata for use in Retrieval-Augmented Generation (RAG) systems, chatbots. Ethiopian History RAG System =========================== This repository implements a Retrieval-Augmented Generation (RAG) system for exploring Ethiopian history, designed for educational and interactive use. The system leverages Wikipedia data, chunking, embedding, and a vector database to provide accurate, context-aware answers to user queries, with a Streamlit-based chat interface. ## Project Description This project enables users (especially kids) to ask questions about Ethiopian history and receive well-formatted, document-grounded answers. The backend uses pre-processed Wikipedia data, chunked and embedded, and stored in a ChromaDB vector database. The frontend is a modern, interactive Streamlit app with persona-based responses. ## Project Structure The directory structure is as follows: ``` ├── code/ │ ├── data_acquisition_wikipedia.py # Wikipedia data acquisition │ ├── data_processing_chunking.py # Cleans and chunks Wikipedia data │ ├── embedding_and_storage.py # Embeds and stores chunks in ChromaDB │ ├── ethiopian_history_streamlit_chat.py # Streamlit chat app │ ├── paths.py # Centralized file paths │ ├── prompt_builder.py # Prompt construction utilities │ ├── save_persona_image.py # Persona image utility │ ├── vector_db_rag_retrieval.py # CLI RAG retrieval and LLM interface │ └── vector_db/ # Vector DB files and chunked data ├── config/ │ ├── config.yaml # App configuration │ └── prompt_config.yaml # Prompt configuration ├── data/ │ ├── ethiopian_history_am.json # Amharic Wikipedia data │ └── ethiopian_history_en.json # English Wikipedia data ├── vector_db/ │ ├── chroma.sqlite3 # ChromaDB database │ ├── ethiopian_history_am_chunked.json # Chunked Amharic data │ └── ethiopian_history_en_chunk …