Modeling of Deep Learning based Intrusion Detection System in Internet of Things Environment

The Internet of Things (IoT) has become a hot popular topic for building a smart environment. At the same time, security and privacy are treated as significant problems in the real-time IoT platform. Therefore, it is highly needed to design intrusion detection techniques for accomplishing security in IoT. With this motivation, this study designs a novel flower pollination algorithm (FPA) based feature selection with a gated recurrent unit (GRU) model, named FPAFS-GRU technique for intrusion detection in the IoT platform. The proposed FPAFS-GRU technique is mainly designed to determine the presence of intrusions in the network. The FPAFS-GRU technique involves the design of the FPAFS technique to choose an optimal subset of features from the networking data. Besides, a deep learning based GRU model is applied as a classification tool to identify the network intrusions. An extensive experimental analysis takes place on KDDCup 1999 dataset, and the results are investigated under different dimensions. The resultant simulation values demonstrated the betterment of the FPAFS-GRU technique with a higher detection rate of 0.9976.