WebClipper

Information finding on the Web is well served by many high-tech search engines. The information found can be collected for sharing and reuse, by individuals as well as the community at large. This will effectively narrow the search domain for web surfers, leading to a reduction in the cognitive load placed on them. This paper reports the development of a personal/community link library building system, called WebClipper, which enables individual and community users to collect index data from Web documents. WebClipper manages the various links of a Web document and extracts its index data to form a virtual digital library. Using WebClipper, a user can collect links and index data from useful Web documents to form his own link library. When searching, he can first search his own link library before going out to navigate the Web. In the community, users can pool their own link libraries for all to use. In this paper, we show an implementation of Web Clipper. It comes with a Web link database management scheme, a Web information clipping technique, and a user interface design. The proposed Web link database management technique allows the system to manipulate link information without archiving physical data, thus eliminating the problem of storage and copyright. The clipping technique allows users to conveniently choose the granularity of the data to be collected; it also carries out automatic keyword extraction and indexing. Database management of link data enables user-defined link descriptions and makes deadlink management easy.