Research On Network Intrusion Detection Based On Fuzzy Clustering Analysis

With the rapid development of Internet technology,more and more network security incidents occur frequently.Today,the threat of information network is not only the original single-host virus infection,but has developed into a host from around the world threat of invasion,and the risk of invasion and diversity more and more complex,so the research of intrusion detection method has very important practical significance.The information in the network exists in the form of data,so the detection of intrusion behavior is actually a process of classifying the network behavior data set.The data set is usually large and requires the methods of data mining.As an important branch of statistics and computational mathematics,clustering analysis is an important method of data mining,and it is more conducive to the realization of intelligent identification because of the unsupervisedproperty.The basic principle of clustering analysis is to divide a data object into a similar class according to a certain attribute.Objects in the same class are similar to each other and differ from objects in other classes.Traditional clustering analysis are usually divided by more stringent standards,a data object can only be divided into a class.However,in the objective reality,some objective objects have the nature of multiple classes,so the division of the real thing need to use the "soft division" method,based on fuzzy mathematics fuzzy clustering algorithm from this produces.Fuzzy clustering analysis method introduces the concept of membership degree in fuzzy mathematics,and theoretically can support the classification of soft division.Fuzzy clustering analysis algorithm has a wide range of applications,can better deal with practical problems.Using the method of fuzzy clustering analysis to excavate the intrusion behavior can get better classification results,more conducive to the identification of the invasion mode.This paper mainly studys the transitive closure of fuzzy clustering algorithm and fuzzy C-means clustering algorithm,andthe application of two algorithms in intrusion detection.Through a series of analysis of intrusion detection data processing experiment,the application of transitive closure method and fuzzy C-mean clustering algorithm is compared.It is proved that the fuzzy C-mean is more efficient and effective in dealing with intrusion detection behavior data,and has better application value.