|
The rapid growth of web pages makes searching and retrieving
information on web efficiently and effectively a critical
problem. While common keyword-based search engines bring
vague and numerous results, we introduce the concept to
query against. URLs which would bound the search results into a
much smaller fraction. As conventional information
filtering services, in which users submit profiles such that
they will be automatically informed of new addtions that may
be of intrest, we expect URL profiles will soon be applied on
similar applications.
In this thesis we propose several index structures for indexing
profiles and algorithms that efficiently match URLs against a
large number of profiles. We also present simulation results
to compare their performance under different scenarios.
|