Improved Dragonfly Optimizer for Intrusion Detection Using Deep Clustering CNN-PSO Classifier

With the rapid growth of internet based services and data generated on these are attracted by attackers to intrude networking information. Based characteristics intruders, many researchers attempted aim detect intrusion with help automating process. Since, large volume is transferred through network, security performance remained an issue. IDS (Intrusion Detection System) was developed prevent intruders secure network systems. The loss still issue because features space grows while detecting intruders. In this paper, deep clustering CNN have been used Meta heuristic algorithms for feature selection preprocessing. proposed system includes three phases such as preprocessing, classification. first phase, KDD dataset preprocessed using Binning normalization Eigen-PCA discretization method. second performed Information Gain Dragonfly Optimizer (IGDFO). Finally, Deep Convolutional Neural Network (CCNN) classifier optimized Particle Swarm Optimization (PSO) identifies attacks efficiently. can be reduced optimization algorithm. We evaluate model NSL-KDD in terms evaluation metrics. experimental results show that achieves better compared existing accuracy, precision, recall, f-measure false detection rate.