Similarity -Based Pattern Recognition for Disease Symptom Extraction and Characterization

Abstract

Neutrosophic Fuzzy Sets (NFS) expand upon classical fuzzy sets in the field of fuzzy set theory
by including measures of truth, indeterminacy, and falsity. This paper thoroughly examines the creation and
assessment of similarity measures for Single-Valued Neutrosophic Fuzzy Sets (SVNFS). The similarity measure is
a crucial metric that quantifies the extent of similarity between two sets. It finds extensive application in various
fields such as pattern recognition, medical diagnosis, and decision-making challenges. Nevertheless, the current
similarity measures of Neutrosophic Fuzzy Sets(NFS) suffer from limited practicality and interpretation, and do
not yield highly reliable outcomes. In order to tackle this issues,We provide a variety of new similarity measures,
including the Hausdorff similarity measure, Membership-grade based similarity measure, and Trigonometric
Hausdorff similarity measure specifically designed for Neutrosophic Fuzzy Sets(NFS). We conduct a comparison
of their performance against existing measures. We verify the efficacy of these approaches by conducting
thorough theoretical research and practical trials, showcasing their suitability in pattern recognition. The
findings demonstrate substantial enhancements in precision and resilience, offering vital tools for academics and
practitioners working with intricate and unpredictable data. The results of our research provide a foundation
for future progress in the Neutrosophic Fuzzy Set theory and its practical use in several areas. 

DOI: 10.5281/zenodo.13926296