Hierarhical Forest Custer and Heuristic Poll Classification For Mining Web Access Logs

Journal Title: International Journal of Engineering and Science Invention - Year 2019, Vol 8, Issue 1

Abstract

Modern web clickstream data comprises multi levels of granularity based on the visitor behavior on its websites, making it cumbersome to scrutinize. Following the overarching idea that the event visualization should provide information regarding the click events at multiple levels of granularity and identifying navigation patterns across several applications, three levels of granularity are identified for analyzing the navigation patterns, namely, web access sequences, set of all users along with log entry. This method is called as Hierarchical Forest Cluster and Heuristic Poll Classification (HFC-HPC). With the objective of mining web access logs and improving the scalability of visualization interface, three parts are designed. They are preprocessing using Localized Sequence Level model, clustering using Certainty Hierarchical Forest and classification via Heuristic Poll. Three identifications are considered during preprocessing namely, user, user session and page view identifications to extract information of interest with minimal time. Followed by the extracted information, Certainty Hierarchical Forest Clustering algorithm is applied with which optimal clusters are said to be generated with minimal computation overhead. Finally, heuristic poll is applied to the generated optimal clusters of applications to address scalability in the context of high level patterns. Experimental results based on web log files illustrate the applicability of the proposed method in achieving better performance than existing methods in terms of computational time, overhead and error.

Authors and Affiliations

B. K. Mathan Nagan, Dr. C. Chandrasekar

Keywords

Related Articles

The Energy and the Work of Engine

We live in the 21. century, we drive cars, we have the thermal and nuclear power plants and it all started with the work of Carnot, who established the principle of creating work of en engine from the supplied warmth. Ho...

Interfacing A 3x8 Matrix Keyboard with the 8086 Microprocessor

All keyboards are available in a variety of sizes. The standard keyboards are the 101 key QWERTY keyboards interfaced to the microprocessor. The small specialized keyboards are also available which contains only 4, 16 an...

Oil Shale Ex-Situ Process - Leaching Study of Spent Shale

During the past decade, significant advancement has been made on various extraction technologies to develop U.S. oil shale resources in an environmentally and economically sustainable fashion. This work has been driven b...

Effect of Heavy Metals on Lysosomal Enzyme Acid Phosphatase Activity of Bivalve L. marginalis…..

Enzymes are referred as biological, catalysts which obey certain general rules. The literature shows investigative lacunae with regard to effects of heavy metals on Acid phosphatase activity of aquatic animals especially...

Download PDF file
  • EP ID EP433236
  • DOI -
  • Views 53
  • Downloads 0

How To Cite

B. K. Mathan Nagan, Dr. C. Chandrasekar (2019). Hierarhical Forest Custer and Heuristic Poll Classification For Mining Web Access Logs. International Journal of Engineering and Science Invention, 8(1), 1-13. https://europub.co.uk/articles/-A-433236