EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Unsupervised Feature Weighting Based on Graph Heat Kernel and Normalized Hilbert–Schmidt Independence Criterion

Long Zou, Tinghua Wang, Yuhong Chen

Entropy · 16.11.2026 · 28 (11) · 1263

DOI: 10.3390/e28111263 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 16.11.2026. Verfasst von Long Zou und 2 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Jiangxi Province of China, Graduate Innovation Special Foundation of Jiangxi Province of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unsupervised feature weighting relies on sample structure to assess feature importance. However, irrelevant features and noise can distort the structural reference, while feature redundancy can cause repeated weighting of similar information. We propose an unsupervised feature-weighting method based on a graph heat kernel and the normalized Hilbert–Schmidt independence criterion (nHSIC), termed GHK-nHSIC-FW, which constructs a label-free structural target using heat diffusion on a nearest-neighbor graph. The method uses nHSIC to measure feature–target alignment and feature redundancy and learns continuous weights with convex quadratic optimization under nonnegative simplex constraints to adjust feature contributions to the weighted Euclidean distance. Experiments on eight public datasets demonstrate that our proposed method can achieve the highest average clustering accuracy (ACC) and normalized mutual information (NMI) among the compared methods, while ablation and subsampling analyses assess component effectiveness and feature importance stability. Synthetic experiments with known informative features examine sample contamination, irrelevant feature injection, correlated feature copies, and diffusion-scale effects under graph perturbations. The results show strong relative performance under sample and irrelevant feature contamination and demonstrate how redundancy control helps retain complementary informative contributions. These results support graph-guided, redundancy-aware feature weighting for unsupervised clustering.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Jiangxi Province of China
  • Graduate Innovation Special Foundation of Jiangxi Province of China

Zitieren

APA

Zou, L., Wang, T., & Chen, Y. (2026). Unsupervised Feature Weighting Based on Graph Heat Kernel and Normalized Hilbert–Schmidt Independence Criterion. Entropy, 28(11), 1263. https://doi.org/10.3390/e28111263

Vancouver

Zou L, Wang T, Chen Y. Unsupervised Feature Weighting Based on Graph Heat Kernel and Normalized Hilbert–Schmidt Independence Criterion. Entropy. 2026;28(11):1263. doi:10.3390/e28111263

BibTeX
@article{Zou2026,
  author = {Zou, Long and Wang, Tinghua and Chen, Yuhong},
  title = {Unsupervised Feature Weighting Based on Graph Heat Kernel and Normalized Hilbert–Schmidt Independence Criterion},
  journal = {Entropy},
  year = {2026},
  volume = {28},
  number = {11},
  pages = {1263},
  doi = {10.3390/e28111263}
}
Kostenlos starten →