An Automation for Agriculture Drought Prediction Using Efficient Machine Learning Algorithms
Ms. Smitha U K1, Dr. Sandeep2
11Ms.Smitha U K, Department of MCA, Navkis College of Engineering, Hassan, Karnataka
2Dr.Sandeep, Associate professor, Department of MCA, Navkis College of Engineering, Hassan, Karnataka
Abstract - Agriculture and its allied sectors form the backbone of India's economy, playing a crucial role in supporting the livelihoods of millions. However, frequent natural calamities such as droughts significantly impact crop yields, leading to severe financial stress for farmers. This project introduces a secure, web-based platform designed to streamline agricultural drought prediction using machine learning techniques. The system focuses on enhancing transparency, traceability, and trust by digitally managing soil-related parameters such as pH level, electrical conductivity, and concentrations of essential nutrients including nitrogen, phosphorus, and potassium. It includes role-based access control for stakeholders such as Application Managers, Agricultural Departments, Farmers, and Researchers. Key features of the platform include real-time data processing, K-Nearest Neighbours (KNN) algorithm implementation, secure data storage using cloud services, and automated communication systems. The use of XML and PDF formats for data handling ensures standardized and secure information exchange. Unlike traditional approaches that rely on static datasets, this model leverages real-time data processing and adaptive learning techniques, enhancing its utility in dynamic agricultural environments. Preliminary testing indicates promising performance in forecasting drought conditions with 91% accuracy, providing a scalable and accessible tool for farmers and agricultural departments. The proposed system holds potential to reduce economic losses and improve resource planning, thereby strengthening resilience against climate-induced agricultural challenges.
Key Words: Data Analytics, Machine Learning, Drought Prediction, K-Nearest Neighbors Algorithm, Supervised Learning Models, Agriculture Sector, Real-Time Processing, Soil Analysis, Climate Adaptation, Agricultural Decision Support.
