Journal of Modeling in Engineering

Journal of Modeling in Engineering

Using web mining to discover rare patterns and monitoring websites

Document Type : Research Paper

Authors
1 Department of Computer Engineering, Jahrom Branch, Islamic Azad University
2 Islamic Azad University
Abstract
This paper examines how to use the concept of entropy to discover rare and unexpected patterns in websites. First, by collecting and pre-processing the log data of ehadish.com website, the required data is prepared for analysis. Then, using the concept of entropy as a measure to measure discontinuity and fluctuation in data, rare and unexpected patterns in the behavior of website users and customers are identified. In the next step, using the k-medoids clustering method, website users are divided into similar groups and the frequency patterns of visits and customer behaviors in each group are examined. Then, using the monitoring technique, suspicious patterns and sudden changes in the website are identified and evaluated, and entropy is used as an important tool to detect these changes. Finally, by producing a management dashboard including reports and analytical results in a graphical and orderly manner, the obtained information is presented to the website administrators. This dashboard allows administrators to compare the website's performance and security with similar examples and take necessary actions to improve their website's security and performance. As an analytical and preventive method, this article can help improve the security and quality of services provided by websites and provide managers with valuable information about user behavior and customer patterns.
Keywords
Subjects


Articles in Press, Accepted Manuscript
Available Online from 27 October 2026

  • Receive Date 16 July 2024
  • Revise Date 30 September 2026
  • Accept Date 10 October 2026