belaymit/amharic-e-commerce-data-extractor

This project extracts entities (products, prices, locations) from Amharic text in Ethiopian e-commerce Telegram channels to create training data for Named Entity Recognition (NER) models Amharic E-commerce Data Extractor # 📗 Table of Contents - 📖 About the Project - 🛠 Built With - Tech Stack - Key Features - 💻 Getting Started - Prerequisites - Setup - Install - Usage - Run tests - Deployment - 👥 Authors - 🔭 Future Features - 🤝 Contributing - ⭐️ Show your support - 🙏 Acknowledgements - ❓ FAQ (OPTIONAL) - 📝 License # 📖 Amharic E-commerce Data Extractor **Amharic E-commerce Data Extractor** is an advanced NLP project that leverages transformer models to extract products, prices, and locations from Amharic Telegram e-commerce channels. The system supports FinTech applications by providing vendor scoring capabilities for micro-lending decisions. ## 🛠 Built With ### Tech Stack Machine Learning PyTorch Transformers Scikit-learn Data Processing Python Pandas NumPy Data Collection Telethon Pyrogram Interpretability SHAP LIME ### Key Features - **Named Entity Recognition for Amharic text** - Extract products, prices, and locations - **Multi-model comparison framework** - XLM-Roberta, DistilBERT, and mBERT - **Model interpretability with SHAP and LIME** - Transparent AI decision making - **Telegram data collection pipeline** - Automated scraping from Ethiopian channels - **FinTech vendor scorecard** - Complete micro-lending risk assessment system - **Business intelligence dashboard** - Vendor analytics and lending recommendations ( back to top ) ## 💻 Getting Started To get a local copy up and running, follow these steps. ### Prerequisites In order to run this project you need: - Python 3.8 or higher - Git - Virtual environment (recommended) ### Setup Clone this repository to your desired folder: ```sh cd my-folder git clone github.com cd amharic-ecommerce-data-extractor ``` ### Install Install this project with: ```sh # Create virtual environment python3 -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate # Install dependencies p …