Privacy Preservation with Machine Learning
P. Bhuvaneswari, Nagender Kumar Suryadevara · 2020
In this chapter, we present various privacy preservation features applied to the Internet of Things (IoT) theme. The IoT framework consists of smart sensing devices, communication systems, and data processing units. The implementations of privacy features across the IoT framework were deliberated. The first section explores the limitations of implementing certain privacy features to the IoT smart devices. The next section focuses on applying privacy features for communication technologies in the IoT system. The third section presents various artificial intelligence data processing techniques for preserving the privacy of data that is originated in the IoT ecosystem. The privacy procedures, such as multi-party computation (MPC), generative adversarial network (GAN), and long short-term memory networks (LTSM), are explained for practical privacy preservation. The fourth section presents the trade-off between the privacy, usage, and efficiency of the presented models. In the end, the machine learning-based privacy preservation techniques and their adaptability to various IoT applications are summarized.