Information concierge for the world wide web
Li Zhao · 2006
With the rapid growth of information on the Web in today's world, a means to combat information overload is critical. Web data extraction systems have been developed to transform, evaluate, manage and present Web documents on behalf of requirements of various applications. Earlier work described in the literature usually focused on a single phase in the data extraction process, especially the generation of rules to transform Web documents; i.e., wrapper induction. They appeared ad-hoc and difficult to integrate; each phase in the data extraction process was disconnected and did not share a common foundation to make the building of a complete system straightforward.