.png){kind=icon}
.png){kind=icon}
Return
Collaborative multi-view fuzzy clustering based on Gaussian mixture model
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neucom.2025.131961.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Current multi-view fuzzy clustering methods are mostly designed based on the strength of prototype-based clustering. In comparison to the prototype-based clustering, Gaussian mixture model (GMM)-based clustering offers greater flexibility in modeling data distributions, making it more applicable to diverse cluster shapes. In this paper, an innovative GMM-based collaborative multi-view fuzzy clustering algorithm is proposed, where a collaborative learning mechanism is designed to facilitate the fusion of multiple views. The valuable knowledge in each view can be learned by other views to guide and improve their own data clustering, and global consistency between views can be guaranteed by adding consensus constraints. Furthermore, to identify inter-view divergence, a self-adaptive learning strategy is established to dynamically adjust the learning rate based on the clustering status of each view. Finally, the maximum entropy regularization is employed to assign optimal weights to each view, emphasizing the importance of the outstanding views. Extensive experiments on real-world multi-view datasets validate the effectiveness of the proposed method in comparison with other traditional multi-view fuzzy clustering algorithms.
Journal
IF:
6.5
Papers:
2.5W
Citations:
6.5W