Anomaly detection of web traffic between IoT Devices
Ali Mohanad Faris AL-Sammarrie, Mesüt Çevik · 2022 International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA) · 2022
There are challenges in anomaly detection and monitoring the data traffic on Internet of Things (IOT) networks which are the results of the state of devices/sensors that's connected to IOT networks. Moreover., protecting and determining the data traffic of devices requires anonymity., another challenge represented by the diversity and heterogeneity of the devices/sensors. Thus., this study proposes a comparison between specific modules which support heterogeneous devices and detect the anomalies data. In this paper., three different ML algorithms (K-Nearest Neighbors., random forest., XGB Classifier) are applied for solving issues related to data traffic management and network monitoring. It is shown that even with the poor data traffic., accuracy between 80% and 89% is achieved for effectively detecting anomality in an IoT network.