The Design And Application Of Network Anomaly Detection Algorithm Based On Affinity Propagation

Along with the advance of Internet technology and the increasing number of network users malicious network behavior presents a complex and diversification trend,combined with the complexity of the Internet itself,heterogeneity,and makes the network security challenges facing more and more serious,so accurate,reliable,abnormal behavior detection method,has the very vital significance.The existing problems in abnormal behavior detection technology exists,separately for each user request processing,ignoring the correlation information between requests,at the same time,the detection means mainly formed through summarizing the existing rules feature library,so the detection effect depends on the feature database updates,for unknown attack detection effect is not ideal,in addition,the detection algorithm is often more complex,difficult to meet the requirements of real-time.In view of the above situation,this paper designs a network abnormal behavior detection method based on near neighbor propagation.The nearest neighbor propagation algorithm is a clustering algorithm based on representative points,which has low time complexity and good clustering effect.Traditional neighbor propagation algorithm can only find the spherical clusters,but on the abnormal behavior detection problem,the distribution of user behavior data is often very complex,traditional neighbor algorithm is difficult to obtain good effect,so in this paper,the traditional neighbor propagation algorithm is improved to optimize,will its application scope to expand to the complex shape clustering problem,to make it breakthrough the limitation of the globular clusters.Then the algorithm is applied to the problem of abnormal behavior detection,and the method based on near neighbor propagation is proposed.The effectiveness of the method is verified by simulation experiments.Based on the above algorithm,this paper designed a based on the spread of the neighbor,abnormal behavior detection system USES the distributed computing and storage technology,designs and realizes an abnormal behavior detection system,the system includes data acquisition,feature extraction,abnormal behavior detection,alarm management,user request query,system management module,network administrators can through these functions to real-time detection of the abnormal behavior of the network information.The system has been tested in the experimental environment,and the results show that the system has a high detection rate and a low false detection rate,which basically meets the security requirements of most websites.