DEVELOPMENT USING BEE-GENT


This section shows development of an electronic catalog retrieval system using the Bee-gent multi-agent framework to exchange catalog data among existing electronic catalog servers.

The architecture of the proposed system is shown in Figure 3. An operation user inputs both search conditions (parts types, parts vendors, create dates, etc.) and retrieval ways (the one to visit the designated servers, the one to visit the servers that store the catalog data of parts provided by the designated vendors , etc.). The operation user sends these data to a mediation agent on a web server and receives the retrieved results.

click to expand
Figure 3: System Architecture of Agent-Based Electronic Catalog Retrieval System

A mediation mobile agent realizes a coordination procedure which includes a migration plan and a query for each server among all electronic catalog servers. The mediation mobile agent migrates to the designated servers according to the search conditions and submits queries to an agent wrapper on the visited server. An agent wrapper agentifies an existing electronic catalog server and invokes it to retrieve catalog data when it receives a retrieval request from a mediation mobile agent.

As shown, the implementations of the electronic catalog servers are distinctly different. In order to encapsulate the implementations , we develop two conversion modules for each agent wrapper. One is from a retrieval request message in ACL to a search API of the wrapped server; the other is from an upload request message in ACL to an upload API of the wrapped server.




(ed.) Intelligent Agents for Data Mining and Information Retrieval
(ed.) Intelligent Agents for Data Mining and Information Retrieval
ISBN: N/A
EAN: N/A
Year: 2004
Pages: 171

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net