The creation of a complex Web site is a thorny problem in user interface de
sign. In this paper we explore the notion of adaptive Web sites: sites that
semi-automatically improve their organization and presentation by learning
from visitor access patterns. It is easy to imagine and implement Web site
s that offer shortcuts to popular pages. Are more sophisticated adaptive We
b sites feasible? What degree of automation can we achieve?
To address the questions above, we describe the design space of adaptive We
b sites and consider a case study: the problem of synthesizing new index pa
ges that facilitate navigation of a Web site. We present the PageGather alg
orithm, which automatically identifies candidate link sets to include in in
dex pages based on user access logs. We demonstrate experimentally that Pag
eGather outperforms the Apriori data mining algorithm on this task. In addi
tion, we compare PageGather's link sets to pre-existing, human-authored ind
ex pages. (C) 1999 Published by Elsevier Science B.V. All rights reserved.