On a two-stage progressive clustering algorithm with graph-augmented density peak clustering

Niu, Xinjian; Zheng, Yunhong; Liu, Wuji; Wu, Chase Q.

doi:10.1016/j.engappai.2021.104566

Cited by 4 publications

(1 citation statement)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DPCG only the grid cell density was calculated by this upgraded method [17]. A two-step clustering method based on graph-enhanced DPC was proposed by Niu et al [18]. Polar coordinates were added to DPC by Li et al in order to lower the increased time complexity [19].…”

Section: Research On Dpc Algorithmmentioning

confidence: 99%

An improved density peaks clustering algorithm use similarity assignment strategy with K-nearest neighbors

Hu,

Feng,

Yang

2024

Preprint

View full text Add to dashboard Cite

Some particular shaped datasets, such manifold datasets, have restrictions on Density Peak Clustering (DPC) performance. The main reason of variations in sample densities between clusters of data with uneven densities is not taken into consideration by the DPC algorithm, which could result in the wrong clustering center selection. Additionally, the use of single assignment method is leads to the effect of assignment errors. To address these problems, this paper creates a new, improved density peaks clustering method use the similarity assignment strategy with K nearest Neighbors (IDPC-SKNN). Firstly, a new method for defining local density is proposed. Local density is comprehensively consider in the proportion of the average density inside the region, which realize the precise location of low-density clusters. Then, using the samples' K-nearest neighbors information, a new similarity allocation method is proposed. Allocation strategy successfully address assignment cascading mistakes and improves algorithms robustness. Finally, based on four evaluation indicators, our algorithm outperforms all the comparative clustering algorithm, according to experiments conducted on synthetic and real world datasets.

show abstract

Section: Research On Dpc Algorithmmentioning

confidence: 99%