Possibilistic Data Analysis and Its Similarity to Rough Sets
This paper is dealing with the upper and lower approximation models for representing the given phenomenon in a fuzzy environment. Based on the given data, the upper and lower approximation models can be derived from upper and lower directions, respectively where the inclusion relationship between these two models holds. Thus, the inherent fuzziness existing in the given phenomenon can be represented by the upper and lower models. The modalities of the upper and lower models have been illustrated in regression analysis and also in the identification methods of possibility distributions. The comparison of the concepts of possibility data analysis and rough sets is shown. A measure similar to the accuracy measure of rough sets is used to clarify the difference between the data structure and the assumed model.
KeywordsConstraint Condition Linear Programming Problem Portfolio Selection Inclusion Relation Possibility Distribution
Unable to display preview. Download preview PDF.
- 2.Guo, P. and Tanaka, H. (1998) Possibilistic data analysis and its application to portfolio selection problems. Fuzzy Economic Review 3 /2, 3–23Google Scholar
- 3.Pawlak, Z. (1982) Rough sets. Int. J. Information and Computer Sciences 11, 341–356Google Scholar
- 4.Pawlak, Z. (1984) Rough classification. Int. J. Man-Machine Studies 20, 469–483Google Scholar
- 9.Tanaka, H. and Guo, P. (1999) Possibilistic Data Analysis for Operations Research. Physica-Verlag, Heidelberg; New YorkGoogle Scholar
- 10.Tanaka, H. and Ishibuchi, H. (1993) Evidence theory of exponential possibility distributions. Int. J. of Approximate Reasoning 8, 123–140Google Scholar
- 12.Tanaka, H., Lee H. and Guo, P. (1998) Possibility data analysis with rough set concept. Proceeding of Sixth IEEE International Conference on Fuzzy Systems 117–122Google Scholar