Federated Learning with Internet of Things for Data Privacy and Security

In recent days, internet connectivity is highly increased due to the mobile technology advancements and internet rapid growth that leads to the generations of vast amount of data and makes Machine Learning’s (ML) training more difficult. In 2018, a data breach is occurred in Facebook and provides some major issues on privacy and security of user’s data that are used to train the ML models. To solve these issues, a recent approach called Federated Learning (FL) is developed by various researchers. Over the traditional ML methods and its classification, recent research mainly focused on the advantages of FL. But certain challenges are occurred in the FL and it must be solved for next stage development. The main aim of this study is to present a solution for the problem of FL with complete assessment in terms of vulnerability. The current development of FL in Internet of Things (IoT) and its applications are studied in this work. The user’s data privacy is protected by different privacy preserving techniques and their importance are also surveyed in this work. The attacks and challenges faced by FL is discussed, where the latest research on the FL’s application with comprehensive investigation is presented in this research work.