.png){kind=icon}
Return
Multi-view clustering algorithm based on sparse regression and structure learning
.png){kind=icon}
PRE
AI
DOI：10.1007/s10586-025-05245-9.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
With the advent of the era of big data, describing the same object from a single perspective can no longer adapt to the development of the times, so it is necessary to conduct multi-perspective clustering research. In this paper, a multi-view clustering method based on structure learning and sparse regression is proposed. A large amount of complex noise in real multi-view data challenges the robustness of existing multi-view clustering methods. In order to make the model robust to outliers, this paper uses the $$l_{2, 1}$$ -norm regression model as a loss function to learn the eigenweight matrix. Many existing multi-view clustering algorithms are limited by high computational complexity. In order to solve this problem, this paper introduces feature selection and uses data clustering after dimensionality reduction. At the same time, in order to fuse the clustering results of each view, we use structural learning to fuse the clustering results of each view into a unified similarity matrix S. The unified clustering results are obtained by spectral clustering at the end, and the method is optimized based on the enhanced Lagrange method. Through theory and experiment, the robustness and good clustering performance of the proposed algorithm are proved.
Keywords:
Multi-view clustering
Structure learning
Sparse regression
Augmenting lagrange
Journal
C
IF:
0
Papers:
691
Citations:
1
Organization
Cited Papers
ShareSaveCentric graph regularized log-norm sparse non-negative matrix factorization for multi-view.png)
SIGNAL PROCESSING
OAAI
SIGNAL PROCESSINGIF3.6
OAAIShareSaveRobust anchor-based multi-view clustering via spectral embedded concept factorizationNEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
ShareSaveShareSave
ShareSave
View more