Sustainable Fertilizer Usage Optimizer for higher yield

The ML-Based Crop Yield Prediction System aims to provide accurate crop yield predictions by utilizing machine learning techniques on agricultural data. The system incorporates various features, such as soil type, weather conditions, rainfall, temperature, fertilizer usage, and irrigation practices, which are known to affect crop growth and productivity. Several machine learning models, including SVM Regressor, Random Forest Regressor, XGBoost Regressor, Convolutional Neural Networks (CNN), Artificial Neural Networks (ANN), Decision Tree, and Gradient Boosting Regressor, are employed to analyze the data and make reliable predictions. These models are trained using historical agricultural data to predict crop yields based on input parameters, helping agricultural planners, farmers, and researchers optimize their crop management strategies.

The system is developed using the Flask framework for the backend, and HTML, CSS, and JavaScript are used to design the front-end interface. The user-friendly application allows users to input various agricultural data points and receive accurate crop yield predictions accurately. By providing a tool that combines data science with agricultural expertise, this project aims to enhance decision-making processes, improve resource allocation, and increase crop productivity. The predictions provided by the system help optimize farming practices, making the process more efficient and sustainable, and ultimately improving food security.

Keywords: Crop yield prediction, Machine learning, SVM Regressor, Random Forest, XGBoost, CNN, ANN, Decision Tree, Gradient Boosting, Flask.