EUVIMEDEUVIMED entdecken
Crossref

KFWAdaBoost-based Embedded Soft Label Learning for Classification

Zhihong Yu, Zongwen Fan, Jin Gou, Cheng Wang

International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems · 2026-09 · 34 (06) · 647-671

DOI: 10.1142/s0218488526500261 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, online erschienen am 14.07.2026. Verfasst von Zhihong Yu und 3 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Gefördert von: Fujian Provincial Science and Technology Plan Guiding Program, Natural Science Foundation of Xiamen, China, Quanzhou City Science & Technology Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

AdaBoost is an adaptive boosting ensemble classifier designed to combine a set of weak classifiers into a stronger ensemble classifier. However, the resource allocation for handling difficult-to-classify samples has not been deeply investigated. To address this, this paper proposes an embedded soft label learning approach called KFWAdaBoost, which aims to provide richer classification information for challenging samples. We first obtain cluster centers through the K-Means Plus Plus (KPP) method. After that, we calculate the similarity between the samples and the cluster centers to provide clustering information for FWAdaBoost. Furthermore, our proposed KFWAdaBoost method is used for more accurate soft label learning on the dataset. The learned soft labels are applied to enhance the information expression of the augmented dataset. By incorporating these soft labels, the approach offers more detailed information for classification, leading to more accurate classification boundaries and higher prediction accuracy. Experimental results show the improvements on a wide range of benchmark classification problems, with a maximum increase of 5.53%. This indicates the effectiveness of the soft label learning approach for classification tasks, highlighting its potential in handling complex and challenging datasets.

Förderung

  • Fujian Provincial Science and Technology Plan Guiding Program
  • Natural Science Foundation of Xiamen, China
  • Quanzhou City Science & Technology Program of China
  • Xiamen Municipal Bureau of Housing

Zitieren

APA

Yu, Z., Fan, Z., Gou, J., & Wang, C. (2026). KFWAdaBoost-based Embedded Soft Label Learning for Classification. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 34(06), 647-671. https://doi.org/10.1142/s0218488526500261

Vancouver

Yu Z, Fan Z, Gou J, Wang C. KFWAdaBoost-based Embedded Soft Label Learning for Classification. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems. 2026;34(06):647-671. doi:10.1142/s0218488526500261

BibTeX
@article{Yu2026,
  author = {Yu, Zhihong and Fan, Zongwen and Gou, Jin and Wang, Cheng},
  title = {KFWAdaBoost-based Embedded Soft Label Learning for Classification},
  journal = {International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems},
  year = {2026},
  volume = {34},
  number = {06},
  pages = {647--671},
  doi = {10.1142/s0218488526500261}
}
Kostenlos starten →