Affiliation:
1. Zhengzhou Tourism College
Abstract
The traditional network intrusion data detection brings a lot of problems, like congestion control, leading to poor clustering effect. An improved intrusion detection algorithm based on ant guide data clustering is proposed, the accuracy of the output probability is determined by the size of the gene-bit random number, updating status categories sufficient statistics to obtain invasion characteristics observation probability and initial probability, and performing cluster center update rules. Simulation results show that the algorithm has better application performance in data clustering and network intrusion detection.
Publisher
Trans Tech Publications, Ltd.
Reference6 articles.
1. Hu Xuekun, Li Jinxia, Song Shuna，et al. Method of Pattern Classification Based on FSVM and RS Theory [J]. Bulletin of Science and Technology, 2010, 26(2): 249-252.
2. Guo Xiaoli, Qu Zhaoyang, Li Xiaodong, et al. Visual Data Mining Model in Power Grid Based on SOM Clustering [J]. Information science，2012, 30(2): 207-209.
3. NOCETI N, ODONE F. Learning common behaviors from large sets of unlabeled temporal series [J]. Image and Vision Computing, 2012, 30(11): 875-895.
4. KHALID S. Motion-based behavior learning, profiling and classification in the presence of anomalies [J]. Pattern Recognition, 2010, 43(1): 173-186.
5. KHALID S. Activity classification and anomaly detection using m-medoids based modelling of motion patterns [J]. Pattern Recognition, 2010, 43(10): 3636–3647.