Journal of Frontiers of Computer Science and Technology ›› 2021, Vol. 15 ›› Issue (12): 2413-2420.DOI: 10.3778/j.issn.1673-9418.2009070

• Theory and Algorithm • Previous Articles     Next Articles

One-Stage Partition-Fusion Multi-view Subspace Clustering Algorithm

ZHANG Pei, ZHU En, CAI Zhiping   

  1. School of Computer, National University of Defense Technology, Changsha 410073, China
  • Online:2021-12-01 Published:2021-12-10

单步划分融合多视图子空间聚类算法

张培,祝恩,蔡志平   

  1. 国防科技大学 计算机学院,长沙 410073

Abstract:

Multi-view subspace clustering has attracted increasing attention for revealing the inherent low-dimension structure of the data. Nevertheless, most existing methods directly fuse the multiple noisy affinity matrices from the original data, and commonly conduct clustering after obtaining a unified multi-view representation. Separating the representation learning from the clustering process can result in a suboptimal clustering result. To this end, this paper proposes a one-stage partition-fusion multi-view subspace clustering algorithm. Instead of directly fusing the noisy and redundant affinity matrices, this paper fuses the more discriminative partition-level information extracted from the affinity matrices. Moreover, this paper proposes a new framework, integrating representation learning, multiple information fusion and final clustering process. The three sub-processes promote each other to serve clustering best. The promising clustering results can lead to better representations and therefore better clustering performance. Consequently, this paper solves the resultant optimization problem through an alternative algorithm. Experiment results on four real-world benchmark datasets show the effectiveness and superior performance of the proposed method over the state-of-the-art approaches.

Key words: multi-view clustering, subspace clustering, data fusion, partition fusion

摘要:

多视图子空间聚类方法因其可以揭示数据内在的低维结构而被广泛关注,但大多数现有的多视图子空间聚类算法直接将多个来自原始数据的充满噪声的相似度矩阵进行融合,并且通常是在得到一致的多视图表示之后再使用[K]均值算法聚类得到最终的结果,这种将表示的学习过程和后续的聚类过程分离的两阶段算法会导致无法得到最优的聚类结果。为了解决这些问题,提出一种单步划分融合多视图子空间聚类算法。该算法不是直接融合具有噪声和冗余信息的相似度矩阵,而是从相似度矩阵中提取出更具有判别性信息的划分级信息进行融合。提出一个新的框架,将表示学习、多视图信息融合以及最后的聚类过程整合在同一框架中。这三个过程彼此促进,好的聚类结果可以引导生成更好的多视图表示,从而得到更好的聚类效果。提出一种有效的轮替优化算法来解决由此得到的优化问题。最后,在四个真实的基准数据集上得到的实验结果可以证明提出方法的有效性以及先进性。

关键词: 多视图聚类, 子空间聚类, 数据融合, 划分融合