New distance metric for fuzzy numbers integrates rescaling

Researchers have introduced the Triangular Fuzzy Rescaling Distance (d_{TR}), a novel metric designed to quantify the distance between Triangular Fuzzy Numbers (TFNs). This new metric uniquely integrates Linear Rescaling directly into its calculation, addressing the challenge of comparing TFNs from attributes with different scales or units without a separate normalization step. The d_{TR} has been formally proven to satisfy metric properties and is bounded, scale-invariant, and origin-invariant, making it suitable for applications in machine learning and multicriteria decision-making. AI

IMPACT This new metric could enhance machine learning algorithms that handle uncertain or heterogeneous data.

RANK_REASON The cluster contains a research paper detailing a new metric for fuzzy numbers. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.LG →

• arXiv
• Hugging Face
• Linear Rescaling
• machine learning
• multicriteria-decision aiding
• Triangular Fuzzy Numbers
• Triangular Fuzzy Rescaling Distance

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →