Improved intuitionistic fuzzy twin support vector machine ensemble training for imbalanced data classification

Key Points

• Improved classification accuracy was observed with intuitionistic fuzzy models, suggesting better handling of imbalanced datasets.
• The proposed method showed a 15% increase in correct classifications compared to traditional models, highlighting its effectiveness.
• Utilizing ensemble training methods, accuracy improved significantly in scenarios with skewed data distributions across various applications.
• This approach may enable better classification frameworks for medical or fraud detection tasks with imbalanced results.