Advertisement

Towards the Handling of Uncertainty in Knowledge Discovery in Databases

  • Sarabjot S. Anand
  • John G. Hughes
  • David A. Bell
Part of the Studies in Fuzziness and Soft Computing book series (STUDFUZZ, volume 39)

Abstract

In this paper we discuss the role of uncertainty in Knowledge Discovery in Databases (KDD) and discuss the applicability of Evidence Theory towards achieving the goal of handling the uncertainty successfully, incorporating it into the discovery process. We claim that Evidence Theory is suitable for representing and handling uncertainty within KDD and present a case for the same. We discuss, EDM, our framework for KDD based on Evidence Theory. EDM consists of representation methods for data and knowledge and operators on the data and knowledge that together form the discovery process. There are several different types of discovery operators within EDM, however, in this paper we limit our discussion to combination operators. We introduce a combination operator called the Proportional Belief Transfer operator and discuss its properties. In particular, we show how it differs from the well-known Dempster-Shafer Orthogonal Sum. We describe the representational correspondence between EDM and the Rough Set model proposing a 2nd degree of “roughness” with the aim of handling missing information in a decision table. We also discuss how the use of the Rough Set model as a basis for Evidence Theory provides a measure for Residual Variation, i.e. the unexmplained varaition in the data set.

Keywords

Residual Variation Mass Function Decision Table Decision Attribute Belief Function 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    R. Agrawal, T. Imielinski, A. Swami (1993), Database Mining. A Performance Perspective, IEEE Transactions on Knowledge and Data Engineering, Special Issue on Learning and Discovery in Knowledge-Based Systems.Google Scholar
  2. [2]
    S. S. Anand, D. A. Bell, J. G. Hughes (1996), EDM: A General Framework for Data Mining Based on Evidential Theory, Data and Knowledge Engineering Journal, Vol. 18, pp. 189–223.CrossRefGoogle Scholar
  3. [3]
    S. S. Anand, D. A. Bell, J. G. Hughes (1995). Handling Uncertainty within Knowledge Discovery in Databases, Internal Report, School of Information and Software Engineering, University of Ulster.Google Scholar
  4. [4]
    D. A. Bell (1993). From Data Properties to Evidence, IEEE Transactions on Knowledge and Data Engineering, 5 (6).Google Scholar
  5. [5]
    J. O. Berger (1985), Statistical Decision Theory and Bayesian Analysis, Springer-Verlag.Google Scholar
  6. [6]
    J. Bernoulli (1713), Ars Conjectandi, Basel.Google Scholar
  7. [7]
    J. Guan, D. Bell (1991), Evidence Theory and its Applications vol. 1, North-Holland.Google Scholar
  8. [8]
    J. Guan, D. Bell (1992), Evidence Theory and its Applications vol. 2, North-Holland.Google Scholar
  9. [9]
    J. Y. Halpern, R. Fagin (1992), Two views of belief: belief as generalized probability and belief as evidence, Artificial Intelligence 54: 275–317.CrossRefGoogle Scholar
  10. [10]
    M. Houtsma, A. Swami (1995). Set-Oriented Mining for Association Rules in Relational Databases, Proc. of the IEEE Conf on Data and Knowledge Engineering.Google Scholar
  11. [11]
    P. Krause, D. Clark (1993), Representing Uncertain Knowledge: An Artificial Intelligence Approach, Intellect Books.Google Scholar
  12. [12]
    H. E. Kyburg Jr. (1987), Representing knowledge and evidence for decision, Uncertainty in Knowledge-Based Systems, B. G. Buchanan and R. R. Yager (eds.) 30–40.CrossRefGoogle Scholar
  13. [13]
    J. F. Lemmer (1986), Confidence factors, empiricism and the Dempster-Shafer theory of Evidence, Uncertainty in Artificial Intelligence L N. Kanal and J. F. Lemmer (eds.), 117–125.Google Scholar
  14. [14]
    G. Paitetsky-Shapiro, W. J. Frawley (1991), Knowledge Discovery in Databases, AAAI, MIT Press.Google Scholar
  15. [15]
    G. Piatetsky-Shapiro (editor) (1993), Working Notes of the Workshop in Knowledge Discovery in Databases, AAAI-93.Google Scholar
  16. [16]
    G. Shafer (1976), A Mathematical Theory of Evidence, Princeton University Press.Google Scholar
  17. [17]
    I. Duntsch, G. Gediga. Roughian (1997): Rough Information Analysis — An Introduction, Internal Report, University of Ulster.Google Scholar
  18. [18]
    A. Skowron, J. Grzymala-Busse (1994). From Rough Set to Evidence Theory, Advances in Dempster-Shafer Theory of Evidence, R. R. Yager, M. Fedrizzi, J. Kacprzyk (ed.), pp. 193–238.Google Scholar
  19. [19]
    Z. Pawlak, J. Grzymala-Busse, R. Slowinski, and W. Ziarko (1995). Rough Sets, Communications of the ACM, Vol. 38, pp. 89–95.CrossRefGoogle Scholar
  20. [20]
    L. A. Zadeh (1965). Fuzzy Sets, Information and Control 8, pp. 338–353.CrossRefGoogle Scholar
  21. [21]
    P. Cheeseman (1985). In Defence of Probability, Proc. IJCAI’85, pp. 1002–1009.Google Scholar
  22. [22]
    R. Kohavi, B. Frasca (1994). Useful Feature Subsets and Rough Set Reducts, Proceedings of the 3`d International Workshop on Rough Sets and Soft Computing.Google Scholar
  23. [23]
    D. Heckerman, A. Mamdani, M. P. Wellman (1995). Guest Editors of Communications of the ACM, Special Issue on Real-World Applications of Bayesian Networks, 38 (3).Google Scholar
  24. [24]
    Suk Kyoon Lee (1992). Imprecise and Uncertain Information in databases: An Evidential Approach, Proc. of the 8th International Conference on Data Engineering, pp. 614–621.Google Scholar
  25. [25]
    I. Duntsch and G. Gediga (1997). Simple Data Filtering in Rough Set Systems, (To appear) International Journal of Approximate Reasoning.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • Sarabjot S. Anand
    • 1
  • John G. Hughes
    • 1
  • David A. Bell
    • 1
  1. 1.Northern Ireland Knowledge Engineering Laboratory School of Information and Software EngineeringUniversity of UlsterNewtownabbey, Co. AntrimNorthern Ireland

Personalised recommendations