EUVIMEDEUVIMED entdecken
Crossref

A Robust Multilabel Feature Selection Approach Based on Graph Structure Considering Fuzzy Dependency and Feature Interaction

Tengyu Yin, Hongmei Chen, Zhong Yuan, Jihong Wan, Keyu Liu, Shi-Jinn Horng, Tianrui Li

IEEE Transactions on Fuzzy Systems · 2023-12 · 31 (12) · 4516-4528

DOI: 10.1109/tfuzz.2023.3287193 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Fuzzy Systems, 2023-12. Verfasst von Tengyu Yin und 6 weiteren Autor:innen. Die Arbeit wurde bisher 97-mal zitiert (Crossref) und nennt 41 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Sichuan Key R&D Project, Natural Science Foundation of Sichuan Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The performance of multilabel learning depends heavily on the quality of the input features. A mass of irrelevant and redundant features may seriously affect the performance of multilabel learning, and feature selection is an effective technique to solve this problem. However, most multilabel feature selection methods mainly emphasize removing these useless features, and the exploration of feature interaction is ignored. Moreover, the widespread existence of real-world data with uncertainty, ambiguity, and noise limits the performance of feature selection. To this end, our work is dedicated to designing an efficient and robust multilabel feature selection scheme. First, the distribution character of multilabel data is analyzed to generate robust fuzzy multineighborhood granules. By exploring the classification information implied in the data under the granularity structure, a robust multilabel $k$-nearest neighbor fuzzy rough set model is constructed, and the concept of fuzzy dependency is studied. Second, a series of fuzzy multineighborhood uncertainty measures in $k$-nearest neighbor fuzzy rough approximation spaces are studied to analyze the correlations of feature pairs, including interactivity. Third, by investigating the uncertainty measure between feature and label, between features, multilabel data is modeled as a complete weighted graph. Then, these vertices are assessed iteratively to guide the assignment of feature weights. Finally, a graph structure-based robust multilabel feature selection algorithm (GRMFS) is designed. The experiments are conducted on 15 multilabel datasets. The results verify the superior performance of GRMFS as compared with nine representative feature selection methods.

Förderung

  • National Natural Science Foundation of China
  • Sichuan Key R&D Project
  • Natural Science Foundation of Sichuan Province
  • Sichuan Science and Technology Achievements Transfer and Transformation Demonstration Project
  • National Science and Technology Council

Zitieren

APA

Yin, T., Chen, H., Yuan, Z., Wan, J., Liu, K., Horng, S., & Li, T. (2023). A Robust Multilabel Feature Selection Approach Based on Graph Structure Considering Fuzzy Dependency and Feature Interaction. IEEE Transactions on Fuzzy Systems, 31(12), 4516-4528. https://doi.org/10.1109/tfuzz.2023.3287193

Vancouver

Yin T, Chen H, Yuan Z, Wan J, Liu K, Horng S, et al. A Robust Multilabel Feature Selection Approach Based on Graph Structure Considering Fuzzy Dependency and Feature Interaction. IEEE Transactions on Fuzzy Systems. 2023;31(12):4516-4528. doi:10.1109/tfuzz.2023.3287193

BibTeX
@article{Yin2023,
  author = {Yin, Tengyu and Chen, Hongmei and Yuan, Zhong and Wan, Jihong and Liu, Keyu and Horng, Shi-Jinn and Li, Tianrui},
  title = {A Robust Multilabel Feature Selection Approach Based on Graph Structure Considering Fuzzy Dependency and Feature Interaction},
  journal = {IEEE Transactions on Fuzzy Systems},
  year = {2023},
  volume = {31},
  number = {12},
  pages = {4516--4528},
  doi = {10.1109/tfuzz.2023.3287193}
}
Kostenlos starten →