Font Size: a A A

Research On Algorithm Of Fuzzy Clustering And Its Application

Posted on:2007-04-01Degree:MasterType:Thesis
Country:ChinaCandidate:J YanFull Text:PDF
GTID:2178360182966687Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Clustering is one of the important tasks in the field of data mining. Fuzzy clustering analysis that introduces the theory of fuzzy sets, provides the capability that be used to deal with real data. And it has been widely used in many fields. In this thesis, principles and general methods of fuzzy clustering are summarized, and analyzed the correlative techniques and present research of clustering algorithms. We discussed typical fuzzy clustering algorithms. The adventages and disadvantages of these algorithms and the problems existing in these algorithms and the prospects of the fuzzy clustering algorithm are discussed.Fuzzy C-means (FCM) clustering algorithm is one of the widely applied fuzzy algorithms at present. But FCM algorithm has some shortcomings. The fuzzy C-means (FCM) clustering algorithm is sensitive to the situation of initialization and easy to fall into the local minimum when iterating. In order to study FCM algorithms systematically and deeply, they are reviewed in this paper based on c-means algorithm, from metrics, entropy, and constraints on membership function or cluster centers.In order to overcome shortcoming of FCM algorithm, in this paper a improved fuzzy C-means clustering algorithm is put forward. The basic idea of the algorithm is modified subjection value by adding weighted value and the optimal choice for parameter of number of clusters c based on cluster validity function.To prove the availability of this improved FCM algorithm, we use the algorithm in two fields: network intrusion detection and web log mining.The intrusion detection is the second floor defence line of the network security. In this paper, we analyze the characteristic of the intrusion detection technique, and bring forward approach of networkintrusion detection based on the improved fuzzy Omeans clustering. The benefit of this approach is that it need not labeled training data sets. Using the data sets of KDD99, the experiment result shows that this approach can detect unknown intrusions efficiently, and increase detection rate of the clustering detection and decrease the false alarms rate.At last, we analyze web log data by using improved fuzzy clustering algorithm to realize web user clustering that is to find the similar user groups according to browsing behaviors and web pages clustering that is to find relate page groups according to the web pages visited by the user. The results of experiment are given to prove the feasibility of using improved fuzzy cluster algorithm in web log data mining.
Keywords/Search Tags:fuzzy C-means (FCM) algorithm, fuzzy cluster, intrusion detection, Web log mining, user clustering, page clustering
PDF Full Text Request
Related items