Machine learning with blockchain for securing agriculture-based applications

Abstract

The security, integrity, and trustworthiness of heterogeneous data have become a burning issue in the rapidly changing smart agriculture environment. This chapter discusses how machine learning (ML) and blockchain technologies can be integrated to ensure the security of agriculture-based applications in the Internet of Multimedia Things (IoMT). It explains how multimodal agricultural data, including sensor measurements, satellite pictures, videos taken by drones, and farmer feedback, can be smartly analyzed with the help of ML and deep learning algorithms and safely stored and shared with the help of blockchain systems. The chapter brings to the fore ML-based methods in detecting anomalies, predicting yields, detecting diseases, and decision support and blockchain capabilities of decentralization, immutability, smart contracts, and traceability. The proposals of the architectural models of ML-blockchain-based agricultural systems are introduced with a focus on secure data exchange, access management, and trust management. Practical use cases such as supply chain monitoring, precision farming, and sustainable resource management are also discussed in the chapter and end with the main challenges, limitations, and future research directions.