.png){kind=icon}
返回
Multi-view clustering based on graph learning and view diversity learning
.png){kind=icon}
PRE
AI
DOI：10.1007/s00371-022-02717-6.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Multi-view clustering is to make full use of different views of the data for clustering. In recent years, many multi-view clustering methods have been proposed. Previous methods usually do not consider the smooth representation of clusters and the diversity of different views simultaneously. Or they are limited to using the average of graphs as the input to the clustering algorithm, ignoring the possible impact of noisy views. Therefore, we propose a multi-view clustering method based on graph learning and view diversity learning. Specifically, in view self-expression learning, manifold learning is added to mine the structural graph information of data and control the geometry of data distribution. After that, view diversity is added, which is used to explore the complementary information of multi-view data and reduce the redundant information of the data. In addition, we also apply an automatic weighting strategy to distinguish different contributions from different views, and generate a consensus graph from multiple similarity graphs. Our work combines graph learning, view diversity learning, and graph fusion in a unified framework for the first time. An alternating iterative method is used to optimize the solution. Experimental results on six data sets show that our model has good clustering performance on different evaluation indicators.
Keyword:
Graph learning
Multi-view clustering
View diversity learning
Graph fusion
期刊
IF:
2.9
论文数:
4.6K
被引数:
6.5K
机构
引用论文
Osteoblast-derived WISP-1 increases VCAM-1 expression and enhances prostate cancer metastasis by down-regulating miR-126.png)
Oncotarget
OAAI
OncotargetIF0
OAAI分享收藏Effect of Educational Program on Knowledge, Skills, and Personal Preparedness for Disasters Among Emergency Nurses: A Quasi-Experimental Study教育项目对急诊护士在灾害方面的知识、技能和个人准备情况的影响：一项准实验研究OAAI
OAAI分享收藏
查看更多内容