Locating Web information using Web checkpoints

Conventional search engines locate information by letting users establish a single Web checkpoint. By specifying one or more keywords, users direct search engines to return a set of documents that contain those keywords. From the documents (links) returned by search engines, user proceed to further probe the WWW from there. Hence, these initial set of documents (contingent upon the occurrence of keyword(s)) serve as a Web checkpoint. Generally, these links are numerous and may not result in much fruitful searches. By establishing multiple Web checkpoints, a richer and controllable search procedure can be constructed to obtain more relevant Web information. This paper presents the design and implementation of permitting multiple checkpoints to facilitate improved searching on the WWW. Web checkpointing is performed as part of the WHOWEDA project.