Identifying of user behavior from server log file

Abstract EN

Due to the increased of information existing on the World Wide Web (WWW), the subject of how to extract new and useful knowledge from the log file has gained big interest among researchers in data mining and knowledge discovery topics.

Web miming, which is a subset of data mining divided into three particular ways, web content mining, web structure mining, web usage mining.

This paper is interested in server log file, which is belonging to the third category (web usage mining).

This file will be analyzed according to the suggested algorithm to extract the behavior of the user.

Knowing the behavior is coming from knowing the complete path which is taken from the specific user.

Extracting these types of knowledge required many of KDD (Knowledge Discovery in Database) steps such as preprocessing, pattern discovery, and pattern analysis.

After that, the complete graph of the visited web will be drawn.

The knowledge discussed in this paper, helps the web designers to improve their web site design and helps to improve their website usability and visitor’s browsing experience by determining related link connections in the website