Discovering interesting navigations on a web site using SAM

Abstract

In this article, a new algorithm called Sequence Alignment Method extended with an Interestingness Measure (SAM(I)) is illustrated for mining navigation patterns on a web site. Through log file analysis, SAMI distinguishes interesting patterns (i.e. unexpected, surprising patterns contradicting with the structure of the web site or direct hyperlinks between web pages) from uninteresting patterns (i.e. expected, known, obvious patterns resulting from the structure of the web site or direct hyperlinks between web pages) and provides information about the order of visited web pages. The algorithm is validated using real data sets of the Music Machines web site http://machines.hyperreal.org, home of musical electronics on the web. Empirical results show that SAMI identifies profiles of visiting behavior, which may be used for web personalization techniques and for optimizing the layout of the web site through structuring of page-links.