Advertisement

Estimation of Tolerance Relation on the Basis of Pairwise Comparisons

  • Leszek Klukowski
Conference paper
Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 95)

Abstract

The methods of estimation of the tolerance relation (overlapping partition) in a finite set on the basis of multiple pairwise comparisons with random errors - developed by the author - are discussed in the paper. Two types of comparisons are considered. The first type (binary) answers the question whether a pair of elements belongs to intersection of two or more subsets. The second type (multivalent) expresses the number of subsets of intersection comprising a pair. The estimates of the relation are determined on the basis of an appropriate discrete programming task. Two estimators are considered: the first one minimizes the sum of differences between relation form and comparisons. The second estimator rests on differences between relation form and medians from comparisons of each pair. The properties of the estimators are based on probabilistic inequalities and simulations.

Keywords

Pairwise Comparison Random Error Preference Relation Relation Form Multivalent Approach 
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.
    David, H.A.: Order Statistics. J. Wiley, New York (1970)zbMATHGoogle Scholar
  2. 2.
    David, H.A.: The Method of Paired Comparisons, 2nd edn. Ch. Griffin, London (1988)zbMATHGoogle Scholar
  3. 3.
    Gordon, A.D.: Classification, 2nd edn. Chapman & Hall/CRC, Boca Raton (1999)zbMATHGoogle Scholar
  4. 4.
    Hoeffding, W.: Probability inequalities for sums of bounded random variables. J. Amer. Statist. Assoc. 58, 13–30 (1963)CrossRefzbMATHMathSciNetGoogle Scholar
  5. 5.
    Klukowski, L.: Some probabilistic properties of the nearest adjoining order method and its extensions. Annals of Operations Research 51, 241–261 (1994)CrossRefzbMATHMathSciNetGoogle Scholar
  6. 6.
    Klukowski, L.: Tests for relation type – equivalence or tolerance – in finite set of elements. Control and Cybernetics 35, 369–384 (2006)zbMATHMathSciNetGoogle Scholar
  7. 7.
    Klukowski, L.: Estimation of tolerance relation the basis of multiple pairwise comparisons with random errors. Control and Cybernetics 36, 443–466 (2007a)zbMATHMathSciNetGoogle Scholar
  8. 8.
    Klukowski, L.: Estimation of the Preference Relation on the Basis of Medians from Pairwise Comparisons in the Form of Difference of Ranks. In: Kurzynski, M., et al. (eds.) Computer Recognition Systems 2. ASC, vol. 45, pp. 232–241. Springer, Heiedelberg (2007b)CrossRefGoogle Scholar
  9. 9.
    Klukowski, L.: Determination of Tolerance Relation - alternative Approach to Intuitionistic and Fuzzy Sets. In: Atanassov, K.T., et al. (eds.) Advances in Fuzzy Sets, Intuitionistic Fuzzy Sets, Generalized Nets and Related Topics. Applications, vol. 2, pp. 85–94. Ac. Publishing House EXIT, IBS PAN, Warsaw (2008)Google Scholar
  10. 10.
    Slater, P.: Inconsistencies in a schedule of paired comparisons. Biometrika 48, 303–312 (1961)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Leszek Klukowski
    • 1
  1. 1.Systems Research Institute Polish Academy of SciencesWarsawPoland

Personalised recommendations