Software Agents for Automating Multiple-Tool Tasks

Chris Toomey and Ray Johnson

A key problem in computer-based information processing is the lack of tools addressing the atttomation of tasks that, span multiple application programs. We are seeking to fill this void by developing Software agent systems through which end users can delegate such tasks to autonomous software agents. We briefly discuss our approach to several key design issues that arise in building such systems: agent knowledge represent, at, ion and reasoning techniques, effective human-agent interfaces, interoperation of external tools, and agent knowledge base extension and maintenance.

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.