A Survey of Web-Usage Mining: Techniques for Building Web-Based Adaptive Hypermedia Systems

Abstract

This chapter discusses Web usage mining techniques that can be applied for building adaptive hypermedia systems. These techniques are used for uncovering hidden patterns within Web access data and then for building the user model that lies in the heart of each adaptive system. Web access data, traditionally stored in the server log files, constitute a rich source of data collected in a non-intrusive way that guards the privacy of users. Several Web usage mining approaches have been proposed for exposing usage patterns, with the most prominent ones being cluster mining, association rule mining, and sequential pattern mining. This chapter provides an overview of the state of the art in research of Web usage mining, and discusses the most relevant criteria for deciding on the suitability of these techniques for building an adaptive Web site. Moreover, the different types of patterns revealed from Web usage mining are correlated with different adaptation aspects.