Can I read Label-Specific Undersampling for Multi-Label Data on EtoBox?

Label-Specific Undersampling for Multi-Label Data by payels is a document available to read on EtoBox.

What is Label-Specific Undersampling for Multi-Label Data about?

The document introduces a novel undersampling method called NaNUML to address class imbalance in multi-label datasets using natural-nearest neighbor principles. This approach allows for label-specific undersampling without parameter optimization, effectively mitigating class imbalance and improving classifier performance across various datasets. Empirical results demonstrate NaNUML

Author

payels

Language

EN