What question did this study set out to answer?

To enhance density peak clustering by improving convergence in unevenly distributed datasets.

March 21, 2026Nonlinear EngineeringOpen Access

DPC-DIST: an improved density peak clustering algorithm based on geometric distribution

View Full Paper

Ask AI

Bookmark

Authors

This algorithm improves clustering outcomes in high-dimensional data, highlighting its practical applications in optimization.

To enhance density peak clustering by improving convergence in unevenly distributed datasets.
Development of the DPC-DIST algorithm incorporating a geometric distribution coefficient.
Evaluation through experiments on six benchmark datasets including Wine and Iris.
Comparison of performance with original DPC, K-means++, and spectral clustering methods.
DPC-DIST outperformed original DPC algorithms in clustering effectiveness.
Achieved a 9.3% increase in Silhouette Coefficient on the Wine dataset.
Obtained a 47.7% improvement in Calinski-Harabasz Score for high-dimensional data.

Jiayong Zhang (2026) studied this question.

View Full Paper

Ask AI

Bookmark

March 21, 2026Nonlinear EngineeringOpen Access

View Full Paper

Ask AI

Bookmark

Authors

This algorithm improves clustering outcomes in high-dimensional data, highlighting its practical applications in optimization.

To enhance density peak clustering by improving convergence in unevenly distributed datasets.
Development of the DPC-DIST algorithm incorporating a geometric distribution coefficient.
Evaluation through experiments on six benchmark datasets including Wine and Iris.
Comparison of performance with original DPC, K-means++, and spectral clustering methods.
DPC-DIST outperformed original DPC algorithms in clustering effectiveness.
Achieved a 9.3% increase in Silhouette Coefficient on the Wine dataset.
Obtained a 47.7% improvement in Calinski-Harabasz Score for high-dimensional data.

Jiayong Zhang (2026) studied this question.

View Full Paper

Ask AI

Bookmark