An efficient web mining algorithm for Web Log analysis: E-Web Miner

This paper is an attempt to apply an efficient web mining algorithm for web log analysis. The results obtained from the web log analysis may be applied to a class of problems; from search engines in order to identify the context on the basis of association to web site design of a e-commerce web portal that demands security. The algorithm is compared with its other earlier incarnation called Improved AprioriAll Algorithm [Tong and Pi-lian, 2005]. It has been shown beyond any doubt through performance analysis that proposed efficient web mining algorithm, E-Web Miner has much better performance in terms of time and space complexity when compared. Even a tracing of the algorithm shows the inherent flow in the Tong and Pi-lian´s algorithm that fails to give correct output. The proposed algorithm, Efficient Web Miner or E-Web Miner can be traced for its valid results and can be verified by computational comparative performance analysis. The number of data base scannings drastically gets reduced in E-Web Miner and the candidate sets are found to be much smaller in stage wise comparison with Improved AprioriAll Algorithm of Tong and Pi-lian, E-Web Miner, thus, is successful to be applied in any web log analysis, including information centric network design.