Spearman-based correlation methods under single-valued neutrosophic sets with application to wheat farmland selection

Paul Augustine Ejegwa, Nasreen Kausar, Tamunokuro Opubo William-West, Hamiden Abd El-Wahed Khalifa, Akbota Nurgaliyeva, Mohammed I. Salman

Scientific Reports, 2026

Full text Open access: Gold Metadata (CSL-JSON)
DOI 10.1038/s41598-026-64472-1
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Wheat remains a main staple crop in Sub-Saharan Africa, yet its cultivation is increasingly endangered by climate unpredictability and environmental uncertainty. The selection of an appropriate farmland for optimal wheat productivity, therefore, involves complex, imprecise, incomplete, and uncertain information. In response to this problem, the proposed study employs the framework of single-valued neutrosophic sets, which has proven effective in modelling imprecision and indeterminacy in decision-making problems. Correlation coefficients under single-valued neutrosophic sets have been deployed to discuss complex selection problems with multi-criteria. Although some single-valued neutrosophic correlation coefficient measures have been established, a number of them lack appropriate statistical justification and reliability. This study introduces new correlation coefficient methods based on Spearman’s approach. The suggested methods are evaluated by comparing them to existing methods, and it is shown that they meet the requirements of the correlation coefficient while providing more accurate and reliable decision outcomes. In particular, based on the numerical experiment in Example 3.1, the two suggested correlation coefficient methods achieve accuracy percentages of 100% and 99.94% in Case I, and 100% and 89.23% in Case II—substantially outperforming the existing methods, which yield notably lower accuracy percentages in both cases. Their applicability is validated through a farmland selection problem aimed at maximizing wheat productivity, using data from a knowledge-based system. The framework is adaptable and can be extended to other complex decision-making scenarios.