birukberhan7-png/ethiopian_crop_chatbot

# 🌿 Ethiopian Crop Recommendation Chatbot ### AI-Powered System for Endemic Plants in Ethiopia --- ## 📋 Project Overview This is a complete AI-powered chatbot that recommends the best **endemic Ethiopian crops** based on soil conditions, climate, altitude, and region — using a **Random Forest ML model** trained on real Ethiopian agro-ecological data (1,500 rows). ### Supported Crops | Crop | Amharic | Type | |------|---------|------| | Teff | ጤፍ | Highland grain | | Enset | እንሰት | Highland staple | | Coffee Arabica | ቡና | Export crop | | Noug | ኑግ | Oilseed | | Barley | ገብስ | Highland grain | | Sorghum | ማሾ | Lowland grain | | Finger Millet | ዳጉሣ | Nutritious grain | | Chickpea | ሽምብራ | Legume | | Linseed | ተልባ | Oilseed | | Maize | በቆሎ | Staple grain | --- ## 🗂️ Project Structure ``` ethiopian_crop_chatbot/ ├── backend/ │ ├── app.py ← Flask server + chatbot engine │ ├── requirements.txt ← Python dependencies │ ├── ml/ │ │ └── train_model.py ← ML training script (run once) │ ├── models/ ← Saved ML models (auto-generated) │ │ ├── crop_model.pkl │ │ ├── soil_encoder.pkl │ │ ├── region_encoder.pkl │ │ ├── crop_encoder.pkl │ │ └── meta.json │ └── dataset/ │ └── ethiopian_crop_recommendation_dataset.xlsx ├── frontend/ │ └── index.html ← Full chatbot UI (pure HTML/JS) └── database/ └── schema.sql ← SQL Server schema + seed data ``` --- ## ⚙️ Installation & Setup ### Prerequisites - Python 3.9 or higher - pip (Python package manager) ### Step 1 — Install Python dependencies ```bash cd backend pip install -r requirements.txt ``` ### Step 2 — Train the ML model (run once) ```bash python ml/train_model.py ``` This reads the Excel dataset, trains a Random Forest classifier, and saves: - `models/crop_model.pkl` - `models/soil_encoder.pkl` - `models/region_encoder.pkl` - `models/crop_encoder.pkl` - `models/meta.json` Expected output: ``` ✅ Accuracy: ~68% …