Machine Learning Applications and Challenges to Protect Privacy in the Internet of Things
Mahadev Anant Gawas, Aishwarya R. Parab, Hemprasad Yashwant Patil · 2022
The Internet of things (IoT) is popularly known as the technology of interconnected devices. IoT aims at improving the quality of life by connecting everything and everyone around the world and imparting varied applications. This pioneering technology does not merely connect the devices over the network but is known to provide diverse features to its users. These features include real-time problem solving, evaluating gathered information, and data stored on the cloud. IoT technology is incorporated into many different types of industries due to its wide range of applications. IoT is generating lots of big data that is further categorized based on time and location dependency, with varying data quality. This data is significant in the decision-making system and plays a very important role in developing smart IoT applications. As we know that IoT is a major source of new data, machine learning will substantially contribute to make smart IoT applications. The growth in IoT has raised some serious concerns, especially in the areas of security and privacy. In recent years, machine learning has paved the way for the IoT. It is a technique that comprises diverse scientific areas. Machine learning utilizes data mining and different methods to find patterns to understand new perceptions from data. Data processing in IoT devices for communications is a significant challenge. Machine learning can play a key role in addressing these concerns. In this chapter, we analyze a range of machine learning techniques that can be used against the challenges presented by IoT. We aim to utilize data in machine learning to get solutions for issues related to privacy in IoT and examine the opportunities and concerns related to it. We categorize these machine learning-based IoT algorithms into those that present systematic solutions to the basic operation challenges in IoT. The potential challenges of machine learning for IoT are also discussed.