Today's Web sites are intricate but not intelligent; while Web navigation i
s dynamic and idiosyncratic, all too often Web sites are fossils cast in HT
ML. In response, this paper investigates adaptive Web sites: sites that aut
omatically improve their organization and presentation by learning from vis
itor access patterns. Adaptive Web sites mine the data buried in Web server
logs to produce more easily navigable Web sites.
To demonstrate the feasibility of adaptive Web sites, the paper considers t
he problem of index page synthesis and sketches a solution that relies on n
ovel clustering and conceptual clustering techniques. Our preliminary exper
iments show that high-quality candidate index pages can be generated automa
tically, and that our techniques outperform existing methods (including the
Apriori algorithm, K-means clustering, hierarchical agglomerative clusteri
ng, and COBWEB) in this domain. (C) 2000 Published by Elsevier Science B.V.
All rights reserved.