mekbib-codes/amharic-ecommerce-data-extractor

Transform messy Telegram posts into a smart FinTech engine that reveals which vendors are the best candidates for a loan. # 🛒 Amharic E-commerce Data Extractor for EthioMart ## 🔍 Project Overview This project develops a **smart FinTech engine** that transforms unstructured Telegram messages into structured business data — helping EthioMart identify the most promising vendors for micro-lending. I build a Named Entity Recognition (NER) system to extract: - 📦 Product Names - 💰 Prices - 📍 Locations and so on ... from **Amharic text, images, and documents** shared across multiple Telegram channels. --- ## 🎯 Project Goals 1. **Automate Telegram Data Extraction** from 5+ e-commerce channels. 2. **Fine-Tune NER Models** (XLM-R, AfroXLMR, mBERT) on labeled Amharic text. 3. **Evaluate & Interpret** model predictions with SHAP/LIME. 4. **Score Vendors** based on business activity, engagement, and pricing. 5. **Support Micro-Lending** through a data-driven vendor analytics engine. ---