Cooperative Multi-Agent Information Gathering

Kelth Decker, Victor M. Lesser, M. V. Nagendraprasad and Thomas Wagner

One goal is to provide seemingly replicated access to known information sources without excessive redundant retrieval of information by multiple agents. A more interesting goal is for the agents to redirect their processing resources and information gathering plans as information is retrieved during query processing. For example, a web page might be retrieved that contains links to new pages that have a high probability of containing useful information but that are not a part of the current retrieval plan. Partial search results may be presented to the user in order to get feedback to focus the remaining search. For example, if the user is looking for information on color printers, the system may discover that there are probably too man), models and would then present the user with a list of features (such as price or print quality) that the user can interact with to narrow the search.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.