Securing Smart Farming Systems Using Multivariate Linear Regression and Long Short-Term Memory

This research aimed to increase the protection offered by intelligent agricultural systems. An algorithmic approach based on machine learning was used to design an architecture to accomplish this objective. The MLR and LSTM, two separate machine learning algorithms, were merged to make the most efficient use of each method. The outcomes of the studies showed that this sector needs some form of irregularity monitoring system. A smart agricultural system constructed by Apphia srl and now being offered for sale collected data over four months. The architecture for detecting anomalies was first developed for use in smart agricultural systems. Still, it has the potential to find use in various other contexts, including smart cities, smart grids, smart health, and smart fabric ecosystems.