Published Paper Details:

Irrigation is planned across the world based on farmers' eye assessment of crops, and as a result, traditional irrigation systems waste around 50% of water. Sprinkle irrigation, drip irrigation, and furrow irrigation are examples of controlled irrigation methods that minimise water waste by 30% to 70%. However, because to the open-loop structure, these techniques fail to maintain precise water content in the soil, resulting in worse crop quality and quantity when soil nutrients are depleted by under or over-irrigation. As a result, input on integrated precision irrigation strategies is required in order to properly use water without compromising crop development. Precision irrigation uses soil moisture, climatic data, and crop type to determine whether or not water is required. As a result, farmers' labour expenses are reduced. In this study, we will determine if agricultural land requires water or not. Using a hybrid machine learning approach known as Hybrid Random Forest and Linear Model, we can determine if we need to apply fertiliser or not. Precision agriculture has recently received a lot of attention as a result of the ever-increasing global population demands for food and water.. As a result, farmers will require water as well as arable land to satisfy this demand. Farmers require a solution that alters the way they work due to the restricted availability of both resources. Precision irrigation is the way to get greater, better, and more lucrative yields with less effort and money. To make better use of water, several machine learning-based irrigation models have been developed. These models are not well adapted to variable climates due to their weak learning capabilities. For precision agriculture, our study offers a deep learning neural network-based Internet of Things (IoT)-enabled intelligent irrigation system. The sensor's data was saved in order to use it to estimate the content and control the water