EUVIMEDEUVIMED entdecken
Crossref

ADMSFI: Anomaly Detection Based on Multisequence Fuzzy Feature Interaction

Zhixuan Deng, Zihao Li, Dayong Deng, Zhonglong Zheng, Gang Li, Tianrui Li

IEEE Transactions on Fuzzy Systems · 2026-06 · 34 (6) · 1804-1817

DOI: 10.1109/tfuzz.2026.3677389 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Fuzzy Systems, 2026-06. Verfasst von Zhixuan Deng und 5 weiteren Autor:innen. Die Arbeit nennt 55 Literaturangaben. Gefördert von: China National Major Research and Development Program of China, National Natural Science Foundation of China, Major Program of the Natural Science Foundation of Zhejiang Province, China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As a core problem in unsupervised learning, anomaly detection focuses on identifying abnormal patterns in datasets, thereby providing support for uncovering potential problems and extracting valuable information. However, most existing methods fail to extract sufficient information in feature interactions when dealing with heterogeneous datasets. To address this challenge, a novel anomaly detection method based on multisequence fuzzy feature interaction is proposed. First, we propose multisequence features based on joint fuzzy information entropy to capture complex feature interactions and to quantify the interdependencies among features. Second, forward and reverse multisequence feature subset pairs are constructed to characterize the correlation between features from different angles, enhancing the accuracy of representing complex interactions in heterogeneous data and improving the ability to identify potential anomalies. Subsequently, an uncertainty measure based on multisequence information fusion is introduced, and anomaly scores are accumulated by incorporating instance weights, thereby ensuring stable detection performance in heterogeneous datasets. Finally, an anomaly detection algorithm based on multisequence fuzzy feature interaction (ADMSFI) is proposed. The experimental results demonstrate that the proposed algorithm ADMSFI significantly outperforms 13 existing algorithms in terms of performance and flexibility in 24 datasets.

Förderung

  • China National Major Research and Development Program of China
  • National Natural Science Foundation of China
  • Major Program of the Natural Science Foundation of Zhejiang Province, China
  • Youth Science Fund of Zhejiang Provincial Natural Science Foundation

Zitieren

APA

Deng, Z., Li, Z., Deng, D., Zheng, Z., Li, G., & Li, T. (2026). ADMSFI: Anomaly Detection Based on Multisequence Fuzzy Feature Interaction. IEEE Transactions on Fuzzy Systems, 34(6), 1804-1817. https://doi.org/10.1109/tfuzz.2026.3677389

Vancouver

Deng Z, Li Z, Deng D, Zheng Z, Li G, Li T. ADMSFI: Anomaly Detection Based on Multisequence Fuzzy Feature Interaction. IEEE Transactions on Fuzzy Systems. 2026;34(6):1804-1817. doi:10.1109/tfuzz.2026.3677389

BibTeX
@article{Deng2026,
  author = {Deng, Zhixuan and Li, Zihao and Deng, Dayong and Zheng, Zhonglong and Li, Gang and Li, Tianrui},
  title = {ADMSFI: Anomaly Detection Based on Multisequence Fuzzy Feature Interaction},
  journal = {IEEE Transactions on Fuzzy Systems},
  year = {2026},
  volume = {34},
  number = {6},
  pages = {1804--1817},
  doi = {10.1109/tfuzz.2026.3677389}
}
Kostenlos starten →