About this document
Dynamic Multi-Label Feature Selection by sizheduan36 is a document available to read on EtoBox.
The paper presents a novel multi-label feature selection method called Weighted Feature Relevancy (WFRFS) that addresses the dynamic uncertainty of label information during feature selection. By categorizing labels into two groups based on their remaining uncertainty, the method utilizes a Relevancy Ratio to evaluate candidate features and improve classification performance. Experimental results demonstrate that WFRFS outperforms six existing multi-label feature selection methods across thirteen real-world
- Author
- sizheduan36
- Language
- EN