The Analysis And Methodologies Of The Existing Problem

1263 Words Sep 15th, 2016 6 Pages
In this chapter, we have discussed about the analysis and methodologies of the existing problem.
4.1 Method Used
After a study of literature review the concluded that the Association rule with Apriori Algorithm is used in combination of Fuzzy c means clustering and it gives result better Accuracy in web page prediction. Also we use Association rule with Apriori algorithm and Markov model in web page prediction then it may also result a highest probability in web page and helped to predict the user access behavior and also decrease the access time of user. So our new proposed methodology has following methods: Fig 4.1.1 Proposed Methodologies
1. Web Log Data The logs files are maintained by the web server by the activity of the client who accesses the web server for a web site through the browser. The information can be written by the site owner, gleaned from the other web sites or other sources or contributed by users. The sample web log dataset are: Fig Sample Log Data
2. Cleaning Log Data In the log data, the entries in the log file for the unwanted view of image, graphics and multimedia by the users are removed. With the important entries, a web log file may consist of certain undesirable rather useless data which has nothing to do with the mining procedure. Therefore, it is imperative to remove those irrelevant entries from the log file. There are three kinds of irrelevant or redundant data needed to clean:…
Open Document