In the field of Web Log Mining, clustering is an important research topic. Fuzzy clustering analysis that introduces the theory of fuzzy sets, provides the capability that be used to deal with real data.FCM algorithm is one of the widely applied fuzzy clustering algorithms at present. But it also there are some shortcomings, the FCM 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 FCM algorithm, from actual distribution of samples and cluster starting centers. An improved FCM algorithm is given. The basic idea is: First of all combine WFCM algorithm with features weighted FCM algorithm, then using a new distance measurement, the final introducing an improved subtraction to the new algorithm for seed selection.At last,we compare the improved FCM algorithm and exiting FCM algorithms using test data sets and real Web log data sets. The experiment proved that improved algorithm has better quality and stable performance. |