Crop Prediction and Optimization Using Hybrid Genetic Algorithm

This study focuses on developing a predictive model for classifying various crops based on key environmental factors such as soil composition and weather conditions. By integrating critical soil parameters, including Nitrogen, Phosphorus, Potassium, and pH, with weather variables like Temperature, Humidity, and Rainfall, the model aims to predict the most suitable crops for specific regions. The approach utilizes a Random Forest Classifier, enhanced through a Genetic Algorithm (GA) for optimizing hyperparameters, thereby improving the model's performance and adaptability to diverse agricultural conditions. The hybrid model is designed to handle complex datasets and offer accurate classifications for multiple crops, including rice, maize, and jute. To ensure transparency and interpretability, the model incorporates Explainable AI (XAI) techniques, such as LIME and SHAP, which provide insights into the decision-making process. This allows users, such as farmers and policymakers, to understand the key factors influencing crop classification. By combining machine learning and optimization techniques, the system offers an effective tool for crop prediction, helping to reduce risks, increase agricultural productivity, and support sustainable farming practices. Ultimately, the model aims to enhance decision-making, mitigate crop failures, and contribute to food security and economic stability in agriculture-dependent regions.