.png){kind=icon}
.png){kind=icon}
Return
SDC: a parameter-free clustering algorithm for incomplete datasets
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2026.113363.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• We propose a parameter-free clustering algorithm for incomplete datasets with missing values. • We devise a single-dimensional strategy that removes imputation and enables decision-graph-based clustering on incomplete data. • We introduce a gravity-inspired mechanism to contract cluster boundaries and enhance separability. • We design a lightweight batch-density estimation scheme that substantially reduces computational complexity. • Extensive experiments show that SDC improves clustering performance by at least 13.7% (NMI), 23.8% (ARI), and 8.1% (Purity) compared with representative baselines. • The source code is publicly available at https://github.com/DJLPKU/single_dimension_clustering .
Keywords:
clustering
incomplete datasets
parameter-free
decision-graph
gravity-inspired mechanism
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W