Machine Learning for Cloud and IoT-Based Smart Agriculture

Predicting soil moisture in Smart Agriculture is essential for deploying Water-Efficient irrigation systems. More significantly, this is a crucial component of the potential Smart Agriculture (SA) system as soil moisture needs to be predicted in real time and linked to the irrigation management system to save water and energy use in agriculture. Various techniques and models have been used to forecast soil moisture. In this study, we looked at the available approaches and opted for machine learning due to the accuracy of its models. Real-world data from a smart farm prototype is used to train and validate the suggested models: Support Vector Regression (SVR) and Long Short-Term Memory (LSTM). These models were then used to predict data related to soil moisture value. Even though our model's prediction accuracy was modest because of the small dataset size, we highly recommend that researchers use it as a blueprint for real-world smart agriculture testbeds and look into machine learning as a promising venue for soil moisture prediction.