Data Mining and Knowledge Engineering

N. Sujatha ¹, K. Iyakutti ²

Affiliations
1 Department of Computer Science, Madurai Kamaraj University, Madurai, Tamil Nadu, IN
2 Department of Microprocessor, Madurai Kamaraj University, Madurai, Tamil Nadu, IN

Abstract

Clustering is one of the important functions in web usage mining. Web usage mining involves application of data mining techniques to discover usage patterns from the web data. Cluster analysis aims at identifying groups of similar objects and, therefore helps to discover distribution of patterns and interesting correlations in large data sets. These methods are not only major tools to uncover the underlying structures of a given data set, but also promising tools to uncover local input-output relations of a complex system. Fuzzy C-means (FCM) is one of the most widely used fuzzy clustering algorithms in real world applications. However there are two major limitations that exist in this method. The first is that a predefined number of clusters must be given in advance. The second is that the FCM technique can get stuck in sub-optimal solutions. In this paper,we have proposed a new framework to improve the web sessions' cluster quality from fuzzy c-means clustering using Genetic Algorithm (GA). Initially the fuzzy c-means algorithm is used to cluster the user sessions. And in the second step, we have proposed a GA based refinement algorithm to improve the cluster quality. The proposed algorithm is tested with web access logs collected from the Internet Traffic Archive (ITA) and shows that refined initial starting points and post processing refinement of clusters indeed lead to improved solutions.

Keywords

Web Usage Mining, Clustering, Fuzzy C-Means, Genetic Algorithm.

Full Text

     

Arun Periasamy ¹, K. Iyakutti ²

Affiliations
1 Madurai Kamaraj University, Madurai, IN
2 School of Physics, Madurai Kamaraj University, Madurai, IN

Abstract

In the day today life the information in World Wide Web is increasing in an explosive way and simultaneously the usage of the Web is also growing in an increasing way and for each and every usage or accessing of the web pages it creates a separate log entry in the web log file and so the log file is also increased correspondingly, From the web log file we get some interesting information about the users previous access sequence. From that interested information it is possible to predict their future access sequence and also personalize the most interested pattern to the users. In this paper we create a model using Tree structure concept, it first scan the collected web log file from that it creates the tree for each and every user and the next step is to traverse the tree using Level-by-Level tree traversal and find out the interested pattern for each user and personalize it to the users in their future access. Using the Level-by-Level tree traversal results the model clusters the uniform interested pattern among the users and it counts the number of clusters and find out the maximum number of interested pattern in the website and also it count the number of interested pattern in each cluster from that the model find out most interested path/pattern of the website during that period of time. From the cluster result it personalized the information to users in their future access. The proposed model is very useful for understanding the behavior of the users, find out the interested object of the website and based upon the results it improving the web site design too, find out the maximum interested pattern of the website, find out the most interested path/pattern of the website. Finally we have done the experimental studies of our proposed model using web log data from a reputed website and prove the efficiency of the proposed model.

Keywords

Web Usage Mining, Clustering, Traversal, Web Personalization.

Full Text