A New Outlier Detection Approach to Discover Low Hit Web Pages Using Sequential Frequent Pattern Mining to Improve Websites Design
The Internet offers huge volume of data to the users and grows rapidly every day. The web server creates log files regarding details about the page, IP address of the user, browser, and operating system used and time/date stamp regarding browsing patterns and this data is mined to extract useful information using web usage mining. The primary objective of this paper is to find the low hit pages of a website from the log files using finding outliers in sequential mining concept. To cater to the need of this objective, a new algorithm named “Detect Anomaly in Sequential Pattern Algorithm (DASPAT)” is proposed. The proposed algorithm creates candidates using Apriori like approach and discovers the unusual browsing behavior of the users, and the detected UBB are treated as outliers. This paper introduces a new approach to find the low hit web pages in tandem to enable the designers to understand how the user browses the site and allow them to redesign the web site
Keywords: Outliers, Sequential mining, Frequent patterns, infrequent patterns, browsing behavior
Download Full-Text
ABOUT THE AUTHORS
K. Subramanian
Research Guide, Department of Computer Science, Govt. Arts and Science, kulithalai, Tamil Nadu, India
S.Vasuki
PhD., Research Scholar, Asst.professor, Department of Computer Applications, J.J. College of Arts and Science(Autonomous), Pudukkottai, Tamil Nadu, India
K. Subramanian
Research Guide, Department of Computer Science, Govt. Arts and Science, kulithalai, Tamil Nadu, India
S.Vasuki
PhD., Research Scholar, Asst.professor, Department of Computer Applications, J.J. College of Arts and Science(Autonomous), Pudukkottai, Tamil Nadu, India
