Open distributed autonomous multi-agent coordination on the Web

Agent technology is becoming more prevalent as the availability of network access, and the demand for the end uses of agents becomes greater. Intelligent agents for information filtering and retrieval applications and tools are being employed in a variety of ways on the Web. A centralized agent for information discovery has usually limited capabilities for finding diverse and distributed information online. The main thrust of the paper is to present a framework that allows distributed adaptive information KODAMA agents to work together to browse and retrieve distributed information based on the preexisting hyperlink structure on the Web and how this community of agents can automatically extract meta-information and cooperate to retrieve online distributed relevant information. We have developed the software architecture, and a working prototype showing the benefits to the Web of interactive architectures based on the coordination of the hyperlink structure on the network. The authors summarize the current results of the project, and discuss some ideas on future work.