| | |
Summary: Predicting Users' Requests on the WWW
I. Zukerman, D.W. Albrecht and A.E. Nicholson
School of Computer Science and Software Engineering
Monash University
Clayton, VICTORIA 3168, AUSTRALIA
phone: +61 3 99055202 fax: +61 3 99055146
fingrid,dwa,annng@csse.monash.edu.au
Abstract. We describe several Markov models derived from the behaviour patterns of
many users, which predict which documents a user is likely to request next. We then present
comparative results of the predictive accuracy of the different models, and, based on these
results, build hybrid models which combine the individual models in different ways. These
hybrid models generally have a greater predictive accuracy than the individual models. The
best models will be incorporated in a system for presending WWW documents.
1 Introduction
Users typically have to wait for information they require from the World Wide Web (WWW).
The eventual aim of this project is to develop a system that reduces a user's expected waiting
time by presending documents s/he is likely to request (Nicholson et al., 1998,Albrecht et al.,
1999). This requires the development of models which can anticipate a user's requests on the
WWW. In this paper, we consider several such models (Sections 4 and 5), and compare their
predictive power and their efficiency in terms of time and space consumption (Section 6). Our
|
