Adaptive fuzzy semi-supervised support vector machine based on sample tightness and its application in credit risk assessment

Key Points

• Improved credit risk assessment outcomes are observed using adaptive fuzzy methods and semi-supervised learning.
• Performance metrics indicate significant gains in accuracy and reliability over traditional methods.
• The method employs a semi-supervised support vector machine to enhance model learning with limited labeled data.
• Findings support the method's potential for practical use in financial risk management and decision-making.