BOTH: Cooperative Automatic Web Navigation and Hierarchical Filtering

Abstract

Due to the rapidly increasing size of the Web Space, the answers returned by the the most common search tools might be at a long distance from the desired documents. Consequently, a great deal of effort must be often carried out by the user to filter less relevant documents pointed out by the search tools answers. This work describes BOTH, a new search tool for World-Wide Web documents that allows the filtering of a specified Web Documents graph. This has been achieved by implementing both a Spider for automatic navigation and a tree-oriented HTML interface for visualizing the retrieved documents. The BOTH Spider retrieves a set of documents linked to one or more initial pages by using a pool of software agents that are ditributed on a LAN of Unix Workstations and are coordinated by a master process. The system explores the relevant documents graph pointed out by the answers returned from a search tool; the results are visualized in a hierarchical way and by following a dynamic approach. The consequence is that the user has a fast way to filter the space of relevant documents without manually browsing through the hypertext. User interface provided by BOTH consists of a query page available on the Web and an HotView page devoted to display the links to the retrieved documents. During the automatic navigation, BOTH Spider stores in a local cache all the pages matching the query and keeps the hypertext structure. Some information about the retrieved documents (abstract, score, etc.) is dynamically written in the HotView page in correspondance of each node of a tree structure. The knowledge of such link structure can therefore be used by the user to identify the most relevant nodes among the ones belonging to the relevant documents graph. After the completion of the automating navigation, these most relevant nodes can be selected in order to decide either to download them to a single HTML answer page or to start a new search from them.

Keywords

WorldWideWeb, Browsing Tools, Information Filtering, Cooperative Agents, Network Computing
Pointers to Full Paper and Conference Presentation
Full Paper Conference Presentation Interactive Version Papers & posters in this theme All papers & posters AusWeb96 Home Page

AusWeb96 Second Australian World Wide Web Conference, Southern Cross University, PO Box 157, Lismore NSW 2480, Australia Email: "ausweb96@scu.edu.au"