SAYED-ZALABIYA/Optimizing-Crop-Selection-Using-Machine-Learning-for-Sustainable-Agriculture-in-Egypt

Machine learning–based crop recommendation system for sustainable agriculture in Egypt (Graduation Project & Published Research). # Optimizing Crop Selection Using Machine Learning for Sustainable Agriculture in Egypt 🎓 Graduation Project 📄 Published Research Paper This repository contains the full implementation of a **machine learning–based crop recommendation system** developed as a graduation project and published in an Egyptian scientific journal. The project applies supervised machine learning techniques to recommend the most suitable crop based on environmental and soil parameters under Egyptian agricultural conditions. --- ## Problem Statement Crop selection in Egypt faces increasing challenges due to: - Climate variability - Water scarcity - Soil degradation Traditional farming practices are insufficient to handle the complexity of modern environmental data. This project proposes a **data-driven decision-support system** to improve crop selection and sustainability. --- ## Dataset & Features Each data sample includes: - Nitrogen (N) - Phosphorus (P) - Potassium (K) - Soil pH - Temperature (°C) - Humidity (%) - Rainfall (mm) Data sources include FAO, Kaggle, and Egyptian Agricultural Research Center datasets. --- ## Methodology ### Models Evaluated - Decision Tree (DT) - Support Vector Machine (SVM) - Linear Regression (LR) - **Random Forest (Proposed Model)** ### Preprocessing & Enhancements - Missing value imputation - MinMaxScaler normalization - Label Encoding - SMOTE for class imbalance - Feature engineering - GridSearchCV with 5-fold cross-validation ### Interpretability - SHAP (SHapley Additive Explanations) used to analyze feature importance and model behavior. --- ## Results The Random Forest model achieved: - **Accuracy:** 100% - **Precision:** 1.00 - **Recall:** 1.00 - **F1-score:** 1.00 - **Cross-validation accuracy:** 99.38% The model demonstrated strong generalization, robustness against overfitting, and interpretability, making it suitable for real-world agricultural advisory systems. --- ## Repository Structure ```text src/ Core ML pipe …