Developing an Automated Distributed Meeting Scheduler

  • Authors:
  • Sandip Sen

  • Affiliations:
  • -

  • Venue:
  • IEEE Expert: Intelligent Systems and Their Applications
  • Year:
  • 1997

Quantified Score

Hi-index 0.00

Visualization

Abstract

Computer networks provide an infrastructure for improving group performance. Tools such as e-mail and shared-file systems help people access and manipulate information. However, they can also impair human performance through overuse or abuse. Introducing intelligent agents into these systems can automate routine organizational tasks, preventing errors produced by the tedious nature of such work. Previously, my colleagues and I studied the usefulness of different heuristic negotiation strategies to solve a precisely defined distributed meeting- scheduling (DMS) problem. We developed analytical, expected-performance models of heuristic strategy combinations and verified them through simulations. In this article, I describe our current project: to design and implement a software system that uses intelligent meeting-scheduling agents that can negotiate with other agents without compromising their user-specified constraints. A small group of researchers in our department runs the prototype system, and we are using the feedback to improve the interface and expand the functionality of the system.