Novel Pythagorean Fuzzy Correlation Measures Via Pythagorean Fuzzy Deviation, Variance, and Covariance With Applications to Pattern Recognition and Career Placement

Abstract

Pythagorean fuzzy set (PFS) is a significant soft computing tool for tackling embedded fuzziness in decision-making. Many computing methods have been studied to facilitate the application of PFS in modeling practical problems, among which the concept of correlation coefficient is very important. This article proposes some novel methods of computing correlation between PFSs via the three characteristic parameters of PFS by incorporating the ideas of Pythagorean fuzzy deviation, variance, and covariance. These novel methods evaluate the magnitude of relationship, show the potency of correlation between the PFSs, and also indicate whether the PFSs are related in either negative or positive sense. The proposed techniques are substantiated together with some theoretical results and numerically validated to be superior in terms of reliability and accuracy compared to some similar existing techniques. Decision-making processes involving pattern recognition and career placement problems are determined using the proposed techniques.