by Han and Chen-Chuan
This article discusses data mining as key technology for bringing intelligence and direction to our Web interactions.
At first they discuss the Web as potential data source for Web mining. Afterwards they elaborate on the following approaches for accessing the Web: (i) keyword based search or topic directories such as DMoz, (ii) querying the deep Web (e.g. Amazon), (iii) random surfing.
The authors think that Web data mining will be important for Web Intelligence to counter (i) the inadequacy of keyword searches, (ii) the limited access to deep Web data and (iii) the restrictions of algorithms such as Page rank. They suggest a multi layered information base as starting point for performing Web mining.