A Kind of Network Intrusion Detection Algorithm Based on Quantum-behaved Particle Swarm Optimization
Qiang Song, Lingxia Liu · DOAJ (DOAJ: Directory of Open Access Journals) · 2013
In order to overcomes the drawbacks of fuzzy clustering methods which are sensitive to the initial values and easily trapped into local minima in intrusion detection algorithm, a hybrid algorithm is proposed based on quantum-behaved particle swarm optimization and semi-supervised kernel fuzzy clustering algorithm. This algorithm can supervise and clustering a few labeled data to generate correct model, use this model to guide lots of unlabeled data to clustering, and enlarge the labeled data set. Those data still cannot be labeled, which are clustered by the kernel fuzzy methods based on quantum-behaved particle swarm optimization, and determine mark types. The simulation of KDD CUP 99 data set is implemented to evaluate the proposed algorithm. Comparing to other algorithms, the result shows the proposed algorithm can obtain the ideal error detection rate and false drop rate in the intrusion detection.