Crop Classification and Yield Prediction Using Robust Machine Learning Models for Agricultural Sustainability

Agricultural sustainability heavily relies on precise crop classification and yield prediction to optimize resource usage and enhance productivity. This study proposes a robust machine learning-based approach leveraging sensor data to support smart farming practices. Key environmental parameters such as soil nutrients (NPK), temperature and humidity (DHT11), and soil moisture are continuously monitored using sensors deployed in the field. The collected data is transmitted to a processing unit where a machine learning model, specifically the Random Forest algorithm, is applied to classify suitable crops—such as maize, paddy, and wheat—based on real-time soil and environmental conditions. This machine learning-driven prediction system assists farmers in making informed decisions regarding crop selection, ultimately improving yield, reducing input waste, and promoting sustainable agricultural practices.