zemawi/Amharic-E-commerce_NER

Fine-tuning large language models (LLMs) for Amharic Named Entity Recognition (NER) to extract key business entities—such as product names, prices, and locations—from text shared on Telegram channels. # Amharic-E-commerce This project builds a platform to support Amharic-language e-commerce on Telegram, focusing on automating the extraction of key business information such as prices, product mentions, and vendor locations. With Telegram becoming a major channel for small businesses in Ethiopia, many vendors operate independently — making it difficult for customers to discover, compare, and interact across vendors. This project aims to solve that by creating a centralized and intelligent system for data extraction and analysis. # Project Objectives * Unify data from multiple Telegram e-commerce channels into a single searchable interface. * Develop a Named Entity Recognition (NER) system tailored for Amharic text. * Extract business-critical entities such as: * Product descriptions * Prices * Store locations * Enable real-time vendor analytics to support features like micro-lending, trend tracking, and performance scoring. # Getting Started Prerequisites Make sure you have the following installed: Python 3.x Pip (Python package manager) Installation Clone the repository:github.com cd Amharic-E-commerce_NER Create a virtual environment and activate it: python -m venv venv source venv/bin/activate # On Windows use `venv\Scripts\activate` Install the required packages: pip install -r requirements.txt