Mining of Different Types of Hidden Knowledge from Log Files

Wajih Abdul Ghani Abdul Hussain, Dr. Hussain k. Al _khafaji

Abstract


Due to the hugeness of information available on the World Wide Web (WWW), extracting novel and useful knowledge from the web
has gained significant attention among researchers in web mining. This type of mining has been used in three particular ways, web content
mining, web structure mining, web usage mining. This paper is related to web usage mining by using the association rules and suggested
algorithms to extract hidden knowledge in the log file.
Extracting these types of knowledge required many of KDD steps such as preprocessing, pattern discovery, and pattern analysis.
After that, the developed Apriori algorithm is adopted to mine the association rules from frequent pages or frequent IPs. The approach discussed
in this paper, helps the system administrator and 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.


Keyword: Data mining, web mining, web usage mining, server log file, association rules, developed Apriori algorithm.


Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v4i4.1658

Refbacks

  • There are currently no refbacks.




Copyright (c) 2016 International Journal of Advanced Research in Computer Science