Title: Web access pattern mining - a new method

Authors: A. Rajimol; G. Raju

Addresses: School of Computer Sciences, Mahatma Gandhi University, Kottayam, Kerala, India; Department of Computer Applications, Marian College, Kuttikkanam P.O. Peermede, Idukki District, Kerala, 685 531, India ' Department of IT, School of Information Science and Technology, Kannur University, Campus P.O. Kannur, Kannur District, 670 567, India

Abstract: An efficient web access pattern mining algorithm, FOL-mine is presented in this paper. The FOL-mine algorithm is based on the projected database of each frequent event and eliminates the need for construction of pattern tree. A data structure, first occurrence list (FOL), is introduced in the proposed algorithm for efficient handling of suffix building. Rebuilding of projection databases is completely eliminated in the new method. Experimental analysis of the algorithms reveals significant performance gain over other web access pattern mining algorithms.

Keywords: sequential access pattern mining; sequential pattern mining; web access pattern mining; web access pattern tree; web log mining; web usage mining; suffix building; access patterns.

DOI: 10.1504/IJWS.2014.066434

International Journal of Web Science, 2014 Vol.2 No.3, pp.140 - 158

Accepted: 20 Mar 2014
Published online: 31 Dec 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article