An Analysis to Find the Efficient Clustering Algorithm for Identification of User Access Pattern

It is established fact that the out of data available on the internet much of it is unnecessary information which may be bug navigation requests, picture request, and incorrect demands. To avoid such situation the proposed logical data cleaning method is capable of cleaning irrelevant data by deducting from the web log records. We have applied four algorithm such as K-Means, Fuzzy C-Means, Divisive hierarchical clustering and Agglomerative hierarchical clustering on a dataset of “Educational Process Mining”, derived by the UCI Repository System for analyzing the distinct outcomes among them, Based on the co-relation coefficient that is Fuzzy silhouette index, Average silhouette width, Divisive coefficient and Agglomerative coefficient.