Credit Scoring and Risk Prediction of Bank Clients Using Correlation-Based Multi-Criteria Group Decision Making in Interval-Valued Complex Spherical Fuzzy Soft Sets

Authors

  • Usman Zulfiqar Lahore Leads University, Lahore, Pakistan
  • Wiyada Kumam Applied Statistics, Department of Mathematics and Computer Science, Faculty of Science and Technology, Rajamangala University of Technology Thanyaburi (RMUTT), Pathum Thani 12110, Thailand
  • Saalam Ali Department of Mathematics, Faculty of Science, King Mongkut's University of Technology Thonburi (KMUTT), 126 Pracha Uthit Rd., Bang Mod, Thung Khru, Bangkok 10140, Thailand
  • Hassan Naseer Department of Mass Communication, Virtual University of Pakistan, Lahore 54000, Pakistan
  • Hamza Naveed Department of Mathematics, Saveetha School of Engineering, SIMATS Thandalam, Chennai-602105, Tamil Nadu, India
  • Rana Muhammad Zulqarnain Jadara Research Center, Jadara University, Irbid 21110, Jordan

DOI:

https://doi.org/10.58715/bangmodjmcs.2026.12.13

Keywords:

interval-valued complex spherical fuzzy soft sets, Hermitian correlation, weighted correlation, MCGDM, credit scoring, risk prediction

Abstract

Credit scoring and risk prediction are essential banking procedures for evaluating financial behaviour before loan approval and protecting institutional stability. This study develops real-valued correlation and weighted correlation coefficients for interval-valued complex spherical fuzzy soft sets (IVCSFSS) and embeds them in a multi-criteria group decision-making (MCGDM) framework. A compact complex-vector representation, Hermitian products, and an outer modulus make the coefficients real, non-negative, symmetric, and bounded in \([0,1]\). The numerical case study prioritizes four credit-assessment policy alternatives against six banking criteria using management and expert evaluations, with all entries screened for interval ordering and spherical admissibility before normalization. The revalidated unweighted scores are \(A_1=0.6949\), \(A_2=0.8790\), \(A_3=0.7337\), and \(A_4=0.8977\); the weighted scores are \(A_1=0.7329\), \(A_2=0.8797\), \(A_3=0.7249\), and \(A_4=0.8928\). The account-duration policy \((A_4)\) ranks first in both formulations, followed by the credit-history policy \((A_2)\). Perturbation tests retain the leading pair, although \(b=1.5\) produces a near-tie reversal \((A_2=0.8915\); \(A_4=0.8914)\). The framework targets decision settings with interval uncertainty, explicit neutrality, phase information, and parameterized soft evaluations.

 

Downloads

Download data is not yet available.

Downloads

Published

2026-09-08

How to Cite

Zulfiqar, U., Kumam, W., Ali, S., Naseer, H., Naveed, H., & Zulqarnain, R. M. (2026). Credit Scoring and Risk Prediction of Bank Clients Using Correlation-Based Multi-Criteria Group Decision Making in Interval-Valued Complex Spherical Fuzzy Soft Sets. Bangmod International Journal of Mathematical and Computational Science, 12(January-December), 289-314. https://doi.org/10.58715/bangmodjmcs.2026.12.13