Central point link learning guided sparse dynamic diagonal embedding for feature selection

Ronghua Shang

doi:10.1016/j.patcog.2025.112233

Central point link learning guided sparse dynamic diagonal embedding for feature selection

2025-08-05

0

PRE

AI

DOI：10.1016/j.patcog.2025.112233

Original

Original request for help

Save

Abstract

En 中文

• This paper proposes a novel unsupervised feature selection method named CPLDE. • Dynamically construct diagonal graph to keep intrinsic relationship between data. • Construct the central point link graph to keep extrinsic connection between data. • The knowledge of central point link guide sparse dynamic diagonal graph embedding. • Design a framework integrated dual graph and maximization the between-class scatter.

Keywords:

CPLDE

unsupervised feature selection

diagonal graph

central point link

dual graph integration

Central point link learning guided sparse dynamic diagonal embedding for feature selection

Abstract

Journal

Organization