Task-based information management

ABSTRACT

Effective collaboration in fast-changing environment can put great dem ands on a collaborator's time. Therefore, information retrieval and filtering tools for these environments should impose as little on that time as possible. Not only should they exclude as many irrelevant documents as possible from those presented to the user (to avoid the time wasted sorting through and reading those documents), they should also minimize the user's effort in characterizing his or her information needs. The goal of the Task-based Information Distribution Environment (TIDE) system is to achieve these objectives by explicitly representing each collaborator's current task and using those representations to deliver documents that meet the information needs implied by those tasks. It does this by treating information gathering as a diagnosis problem, in which the situation (i.e., the current state of beliefs about various questions related to a task) leads probabilistically to test that will provide the most evidence toward reaching a diagnosis (i.e., a description of the documents most likely to be useful to that task). It encodes tasks as nodes in a Bayesian network, and computes document descriptions based on the probabilistic relationship among tasks and their corresponding information requirements.

REFERENCES

• 1: Nicholas J. Belkin , W. Bruce Croft, Information filtering and information retrieval: two sides of the same coin?, Communications of the ACM, v.35 n.12, p.29-38, Dec. 1992 [doi>10.1145/138859.138861]
• 2: Chris Buckley, Implementation of the SMART Information Retrieval System, Cornell University, Ithaca, NY, 1985
• 3: Friedman, N. and Goldszmidt, M. 1996a. Building classiers using Bayesian networks. In Proceedings of the Thirteenth National Conference on Articial Intelligence (AAAI'96) (1996).
• 4: Friedman, N. and Goldszmidt, M. 1996b. Learning Bayesian networks with local structure. In Proc. Twelfth Conference on Uncertainty in Articial Intelligence (UAI'96) (1996).
• 5: McGuinness, D. 1998. Ontological issues for knowledge-enhanced search. In Proceedings of Formal Ontology in Information Systems (1998).
• 6: Judea Pearl, Probabilistic reasoning in intelligent systems: networks of plausible inference, Morgan Kaufmann Publishers Inc., San Francisco, CA, 1988
• 7: Gerard Salton , Michael J. McGill, Introduction to Modern Information Retrieval, McGraw-Hill, Inc., New York, NY, 1986
• 8: Yan, T. W. and Garcia-Molina, H. 1995. SIFT|a tool for wide-area information dissemination. InProceedings of the 1995 USENIX Technical Conference (1995), pp. 177-186.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval
          Subjects: Information filtering

Additional Classification:
  G. Mathematics of Computing
  G.2 DISCRETE MATHEMATICS
      G.2.2 Graph Theory
          Subjects: Network problems

  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval
          Subjects: Retrieval models
  H.5 INFORMATION INTERFACES AND PRESENTATION (I.7)

General Terms:
Design, Human Factors, Reliability