Improving the Growth of Jowar Millets in the Region of Uttar Pradesh Using Machine Learning and IoT

Agriculture is the main driver of the Indian economy, and technology is transforming conventional farming to increase productivity while reducing resource waste. Water scarcity is the biggest problem facing the globe today, mostly affecting the agricultural system. Consequently, a unique approach is taken while implementing the technology to increase agricultural output. This study aims to develop a smart irrigation system that utilises machine learning enabled by the Internet of Things (IoT) to automatically water plants and notify the user when needed. We have examined the jowar millets' development pattern in the Rampur region of Uttar Pradesh when there is a water shortage. Though it is grown south of Vindhyas, we have used our suggested model in this area because it requires little water. The gathered data is received by a cloud-based server, which saves it, interprets it, and applies machine learning to make irrigation recommendations. Field sensors are used in this project and the data they collect is stored on a cloud server. When deciding how much water to deliver, a decision-making tool called SVM is employed. SVM considers both soil moisture content and ambient temperature while determining how much water to disperse. With an accuracy rate of 99.8%, the SVM classifier is the best; it is closely followed by the RVM classifier, which has an accuracy rating of 98.8%. By contrasting each recommended strategy with earlier research, its efficacy is verified. Based on recognised performance indicators, the experimental results show that the recommended strategy performs better than comparable approaches.