.png){kind=icon}
.png){kind=icon}
Return
Multi-view Ensemble Feature Selection via SemiDefinite Programming
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ejor.2025.07.014.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposes MEFS framework integrating view generation and selection via MaxCut optimization. • Introduces pairwise diversity metric using between-class scatter matrices for multi-view learning. • Develops SDP relaxation with inequality constraints for robust and scalable feature selection. • Validated on 10 datasets with superior accuracy, stability, and computational efficiency. • Ablation studies confirm synergistic effect of view generation and selection modules.
Keywords:
MEFS framework
multi-view learning
MaxCut optimization
feature selection
SDP relaxation
Journal
IF:
6
Papers:
2.2W
Citations:
6.4W
Organization
No organization information available