A novel Chaotic Flower Pollination-based intrusion detection framework

摘要: With the rise of network on handheld devices, security has become critical issue. Intrusion detection system is used to predict intrusive packets network; two-step procedure been intrusion, i.e., feature selection and then classification. Firstly, unwanted expandable features in data lead classification problem which affect decision capability classifiers, so we need optimize technique. Feature technique this paper based correlation information known as correlation-based (CFS). In paper, CFS’s search algorithm implemented using Chaotic Flower Pollination Algorithm (CFPA) that logically selects most favorable for referred CFPA-CFS. Further, hybridization CFPA support vector machine classifier named CFPSVM. Finally, novel IDS framework uses CFPA-CFS CFPSVM sequence intrusion. performance proposed evaluated two intrusion evaluation datasets, namely KDDCup99 NSL-KDD. The results demonstrate contributes more achieve better accuracy compared with state-of-the-art methods.