Decision-Centric Information Monitoring

  • Authors:

  • Len Seligman;Paul Lehner;Ken Smith;Chris Elsaesser;David Mattox

  • Affiliations:

  • The MITRE Corporation, McLean, VA 22102. seligman@mitre.org;The MITRE Corporation, McLean, VA 22102. plehner@mitre.org;The MITRE Corporation, McLean, VA 22102. kps@mitre.org;The MITRE Corporation, McLean, VA 22102. chris@mitre.org;The MITRE Corporation, McLean, VA 22102. mattox@mitre.org

  • Venue:
  • Journal of Intelligent Information Systems
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

We are interested in information management for decision supportapplications, especially those that monitor distributed,heterogeneous databases to assess time-critical decisions. Users ofsuch applications can easily be overwhelmed with data that may changerapidly, may conflict, and may be redundant. Developers are facedwith a dilemma: either filter out most information and risk excludingcritical items, or gather possibly irrelevant or redundantinformation, and overwhelm the decision maker. This paper describes asolution to this dilemma called decision-centric informationmonitoring (DCIM). First, we observe that decision support systemsshould monitor only information that can potentially change somedecision. We present an architecture for DCIM that meets therequirements implied by this observation. We describe techniques foridentifying the highest value information to monitor and techniquesfor monitoring that information despite autonomy, distribution, andheterogeneity of data sources. Finally, we present lessons learnedfrom building LOOKOUT, which is to our knowledge the firstimplementation of a top-to-bottom system performing decision-centricinformation monitoring.