.png){kind=icon}
Return
Multi-view clustering algorithm based on feature learning and structure learning
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neucom.2024.128138.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Multi-clustering aims to cluster heterogeneous information from different perspectives, which is an important research direction in Multi-view learning. Many clustering algorithms based on multi-view data usually have the following two problems: (1) high-dimensional data leads to high computing cost, and data noise leads to poor clustering performance. (2) It cannot effectively carry out structure learning and integrate heterogeneous information of each view. In order to address these issues, this work suggests a multi-view clustering algorithm (FS-MVC) based on feature learning and structure learning. For each view, feature learning comes first. The data after dimensionality reduction is used for clustering. The projection matrix is introduced to reduce the dimensionality of each view of the original data. Second, a brand-new structure learning technique is developed that may efficiently utilize the diverse information present in each view and produce the best clustering outcomes. The clustering results on eight multi-view data are compared with those of nine advanced multi-view clustering methods in order to confirm the performance of the proposed approach.
Keywords:
Multi-view clustering
Feature learning
Structural learning
Cluster labels
Journal
IF:
6.5
Papers:
2.5W
Citations:
6.5W
Organization
No organization information available
Cited Papers
ShareSaveShareSaveRobust anchor-based multi-view clustering via spectral embedded concept factorization.png)
NEUROCOMPUTING
NEUROCOMPUTINGIF6.5
PREAI
ShareSaveShareSave
ShareSave
View more