.png){kind=icon}
返回
Consistency-oriented clustering ensemble via data reconstruction
.png){kind=icon}
PRE
AI
DOI：10.1007/s10489-024-05654-0.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
The study highlights that using different distance measures on the same dataset leads to varying clustering results, making the choice of distance measure a challenge when prior knowledge is lacking. To address this issue, a consistency-oriented clustering ensemble via data reconstruction is developed. This approach eliminates the need to select a specific distance measure and achieves higher consistency between the clustering ensemble and base clusterings while maintaining superior clustering performance. First, the base clustering is generated via the clustering with different distance measures and a consistency definition is introduced in the proposed method. Then the ensemble process updates the weights of base clusterings to ensure they reach the consistency. At the same time data reconstruction process is integrated into the ensemble process to guarantee a high convergence rate and efficient clustering. Finally, the clustering ensemble result is achieved with the higher consistency measure and improved clustering performance by balancing both factors. In the experiment, the effectiveness of the proposed method is verified and the specification of the parameters is advised through the various experimental outcomes.
Keyword:
Clustering ensemble
Consistency
Data reconstruction
Similarity matrix
Distance measure
期刊
IF:
3.5
论文数:
7.6K
被引数:
1.7W
机构
暂无机构信息
引用论文
Enhancing diversity and robustness of clustering ensemble via reliability weighted measure.png)
APPLIED INTELLIGENCE
APPLIED INTELLIGENCEIF3.5
PREAI
分享收藏
分享收藏
查看更多内容