Comparative Analysis Of Security And Privacy Technique For Federated Learning In IOT Based Devices

The field of machine learning has been seen as a major development in the last few years. Many new algorithms and many new methods have been put forward by various researchers in this domain. Before the COVID-19 pandemic, things were done manually but after this situation, the culture of working from home has been started at almost every organization except few a necessary government organizations which include healthcare and other emergency services. Online work involves a lot of data transfer and hence it is demanded new development in machine learning and this learning emerged as one such development. Federated learning enables multiple devices to build a common machine learning model without sharing data which helps in providing better data privacy because training data are not transmitted to a central server. Federated learning is also known as collective learning where we train the algorithms across various devices with the help of decentralized data samples without the involvement of actual data. In this paper, the authors will provide various use cases, as well as a comparative study of various federated learning frameworks. This paper will provide in-depth knowledge as well as future research directions in the field of federated learning.