Platform Utilizing Others' Behavior Data to Detect Anomalous Operation Hiding Private Information

As the number of IoT consumer electronics is increasing, cyberattacks to IoT devices are increasing. In particular, operating IoT devices by attackers make users feeling unsafe and may harm users physically. Therefore, we have proposed a method to detect anomalous operations by learning users' behavior. However, this method misdetects many legitimate operations if a sufficient amount of data on the users' behavior. One approach to avoid misdetections even if a sufficient amount of data cannot be obtained from each user is to use the data collected from the others. But users do not want to share their private information with others. In this paper, we proposed an anomaly detection platform that utilizes the dataset of similar users without sharing their private information.