Correlation coefficient between fuzzy data

Author

20.1001.1.27174409.1397.1.1.6.7/DOR

Abstract

Correlation is one of the most important indicators that is widely used in various fields and is an important measure in data analysis. Since much of the actual data may be inaccurate, it seems necessary to expand the concept of correlation in the inaccurate environment. In this paper, we first study a criterion for calculating the correlation coefficient between fuzzy numbers based on the classical definition of correlation coefficient and the principle of expansion, in which the mathematical programming approach is used to calculate the degree of membership. Then, another method for determining and calculating the -drastic of the exact T-based algebraic functions of the membership function using it is examined, which does not depend on mathematical programming.

Keywords

References

[1] طاهری م (۱۳۷۵) آشنایی با نظریه ی مجموعه های فازی، انتشارات جهاد دانشگاهی مشهد.
 
[2] Chachi, J., Taheri, S.M., (2016), Multiple Fuzzy Regression Model for Fuzzy Input Out- put Data, Iranian Journal of Fuzzy Systems, Vol 13,pp 63-78.
 
[3] Hong D. H., (2006), "Fuzzy measures for a correlation coefficient of fuzzy numbers under Tw (the weakest t-norm)-based fuzzy arithmetic operations", Information Sciences, Vol 176, pp 150-160.
 
[4] Liu, S. T., Kao, C.,(2002), "Fuzzy measures for correlation coefficient of fuzzy numbers", Fuzzy Sets and Systems, Vol 128, pp 267–275.

Files

History

  • Receive Date: 14 December 2018
  • Revise Date: 21 January 2019
  • Accept Date: 21 January 2019

Share

How to cite

Statistics