Association Rule Mining among web pages for Discovering Usage Patterns in Web Log Data

Main Article Content

L. Mohan
T.Venu Gopal

Abstract

The objective of this project is to find the associations among different web pages in a web log file of a website, and then divide
the page of each user accessed, who is likely to visit the Web site more than once and create individual sessions. This can be done by computing
the association rules in relation to the web pages by using the information from the web log file maintained by its web server. Such information
containing association rules gives a broad view of user sequences. It can be extended to any field of activity that involves large bulks of data.
This paper is concerned with web usage mining of web mining. The server maintains the information of usage data sequences done by different
users. The web log file has the information about the different pages that have been accessed by the users and the secondary information about
each URL. Considering that information, in order to perform an association rule analysis, which is also known as market basket analysis, there’s
the need to define the basket; in the web environment this is not as clear as in a real supermarket. A phase of transaction identifiers is needed.
Finally, after the application of one of these methods, the transaction file is created, the basket entities are defined and the discovery process of
association rules can go on with the application of an Association Rule algorithm such as the Apriori one. To find the confidence among the web
pages accessed, the user gives a user specified threshold value such as minimum support and minimum confidence values, which determines the
probability of accessing a web page when a set of web pages are accessed.

Downloads

Download data is not yet available.

Article Details

Section
Articles