The Web has so fax been incredibly successful at delivering information to various groups of people. Information sources over the WWW contain a large amount of data organized according to different interests and values. It is important to enable people to obtain their interested information in a simple and effective manner. Meanwhile, there is also an urgent need to make information from the WWW accessible to applications, in order to offer automation, inter-operation and Web-awareness among services. To achieve these goals, information from Web sources need to be extracted automatically according to users' interests. Such automatic extraction is usually performed by specialized programs, called wrappers. However, the creation of wrappers often requires users to have in-depth knowledge of relevant technologies and is difficult, slow and tedious for ordinary users.

145 p.