dani6566/Ethiopian-Medical-Business-Data-Warehouse-Telegram-Scraping-Object-Detection

# Ethiopian Medical Businesses Data Warehouse This project focuses on building a **Data Warehouse** for storing and managing data related to Ethiopian medical businesses, primarily sourced from **Telegram channels**. It involves data scraping, cleaning, transformation, and object detection (using YOLO), with a backend developed using **FastAPI**, **SQLAlchemy**, and **Pydantic** for API exposure and data management. ## Features - **Data Scraping**: Collect business-related data from Telegram channels. - **Object Detection**: YOLO is used to detect and extract relevant images (e.g., business logos). - **Data Cleaning & Transformation**: Prepares raw data for analysis and storage. - **Data Storage**: Uses PostgreSQL as the primary data warehouse. - **API Access**: RESTful APIs built with FastAPI for easy data retrieval. ## Tech Stack - **Programming Language**: Python - **Database**: PostgreSQL - **Backend**: FastAPI - **ORM**: SQLAlchemy - **Data Transformation**: DBT (Data Build Tool) - **Object Detection**: YOLO ## Setup Instructions ### 1. Clone the Repository ```bash git clone github.com cd ethiopian-Medical-Business-Data-Warehouse-Telegram-Scraping-Object-Detection ``` ## Future Improvements -**Add advanced data analytics capabilities.** -**Enhance object detection to classify more medical business entities.** -**Broaden data sources beyond Telegram channels.** ## License This project is licensed under the MIT License.