A new graph-based clustering method with dual-feature regularization and Laplacian rank constraint

Hengdong Zhu, Yingshan Shen, Choujun Zhan, Fu Lee Wang, Heng Weng, Tianyong Hao

Abstract: Highlights•A novel graph-based clustering method called GC-DRRC is proposed.•The dual-feature regularization learns the local manifold structure.•The Laplacian rank constraint ensures that the graph has C connected components.•Comprehensive experiments show that GC-DRRC achieves state-of-the-art performance.