Decision-Centric Information Monitoring

We are interested in information management for decision support applications, especially those that monitor distributed, heterogeneous databases to assess time-critical decisions. Users of such applications can easily be overwhelmed with data that may change rapidly, may conflict, and may be redundant. Developers are faced with a dilemma: either filter out most information and risk excluding critical items, or gather possibly irrelevant or redundant information, and overwhelm the decision maker. This paper describes a solution to this dilemma called decision-centric information monitoring (DCIM). First, we observe that decision support systems should monitor only information that can potentially change some decision. We present an architecture for DCIM that meets the requirements implied by this observation. We describe techniques for identifying the highest value information to monitor and techniques for monitoring that information despite autonomy, distribution, and heterogeneity of data sources. Finally, we present lessons learned from building LOOKOUT, which is to our knowledge the first implementation of a top-to-bottom system performing decision-centric information monitoring.

[1]  Gio Wiederhold,et al.  Mediators in the architecture of future information systems , 1992, Computer.

[2]  Inderpal Singh Mumick,et al.  Maintenance Of Materialized Views , 1999 .

[3]  Gang Zhou,et al.  A framework for supporting data integration using the materialized and virtual approaches , 1996, SIGMOD '96.

[4]  Rob Golding.  Things every update replication customer should know (abstract) , 1995, SIGMOD '95.

[5]  Arnon Rosenthal,et al.  Distributed Situation Monitoring: Issues and Architectures , 1996, VIEWS.

[6]  Calton Pu,et al.  CQ: a personalized update monitoring toolkit , 1998, SIGMOD '98.

[7]  Michael Stonebraker,et al.  The design of POSTGRES , 1986, SIGMOD '86.

[8]  Richard E. Neapolitan,et al.  Probabilistic reasoning in expert systems - theory and algorithms , 2012 .

[9]  Larry Kerschberg,et al.  A Mediator for Approximate Consistency: Supporting “Good Enough” Materialized Views , 2004, Journal of Intelligent Information Systems.

[10]  Dennis McLeod,et al.  A federated architecture for information management , 1985, TOIS.

[11]  Rob Goldring.  Things every update replication customer should know (abstract) , 1995, SIGMOD 1995.

[12]  A. Prasad Sistla,et al.  Temporal conditions and integrity constraints in active database systems , 1995, SIGMOD '95.

[13]  Per-Åke Larson,et al.  Updating derived relations: detecting irrelevant and autonomously computable updates , 1986, VLDB.

[14]  Scott A. Musman,et al.  Constructing Belief Networks to Evaluate Plans , 1994, UAI.

[15]  Richard E. Neapolitan,et al.  Probabilistic reasoning in expert systems - theory and algorithms , 2012 .

[16]  Carole D. Hafner,et al.  The State of the Art in Ontology Design: A Survey and Comparative Review , 1997, AI Mag..

[17]  Shlomo Zilberstein,et al.  Value-driven information gathering , 2002 .

[18]  Inderpal Singh Mumick,et al.  Updating Derived Relations: Detecting Irrelevant And Autonomously Computable Updates , 1999 .

[19]  Inderpal Singh Mumick,et al.  Deriving Production Rules For Incremental View Maintenance , 1999 .

[20]  Howard Raiffa,et al.  Decision analysis: introductory lectures on choices under uncertainty. 1968. , 1969, M.D.Computing.

[21]  Clark D. French,et al.  “One size fits all” database architectures do not work for DSS , 1995, SIGMOD '95.

[22]  Arnon Rosenthal,et al.  A Metadata Resource to Promote Data Integration , 1999 .

[23]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[24]  Eric Simon,et al.  Promises and Realities of Active Database Systems , 1995, VLDB.

[25]  Surajit Chaudhuri,et al.  Maintenance of Materialized Views: Problems, Techniques, and Applications. , 1995 .