EUVIMEDEUVIMED entdecken
Crossref

PLCDroid: enhancing android malware detection by mitigating pseudo-label noise in the presence of concept drift

Lingyu Qiu, Zhen Liu, Bitao Peng, Jingjing Zhang, Ning Liu, Zhihao Wu, Ruoyu Wang

The Computer Journal · 14.07.2026 · 69 (7) · 1218-1236

DOI: 10.1093/comjnl/bxag021 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Computer Journal, online erschienen am 24.03.2026. Verfasst von Lingyu Qiu und 6 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 55 Literaturangaben. Gefördert von: Guangdong Basic and Applied Basic Research Foundation, China, Science and Technology Program of Guangzhou, China, Graduate Research and Innovation Program of Guangdong University of Foreign Studies.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Due to the continuous evolution of Android malware, machine learning-based malware detection systems face the challenge of performance degradation. To address this issue, active learning has been employed to retrain models with new labeled data. Traditionally, active learning relies on ground-truth labels, which are time-consuming to obtain. Although leveraging model-predicted pseudo-labels for model retraining offers a cost-effective alternative, incorrect pseudo-labels may lead to model self-contamination. To alleviate the annotation overhead during model retraining and mitigate the detrimental effects of erroneous pseudo-labels on active learning performance, we introduce a novel framework, PLCDroid. The framework incorporates a label correction mechanism when using pseudo-labels for model retraining. Specifically, we present a pseudo-label type recognition method (PTR) based on model uncertainty and confidence to identify incorrect pseudo-labels. On the basis of PTR, we design fine-grained correction strategies to refine pseudo-labels. Consequently, the proposed method mitigates pseudo-label errors, thereby improving malware detection performance under concept drift. Experimental results over a decade-long period demonstrate the effectiveness of our approach. In the retraining task, leveraging corrected pseudo-labels leads to a substantial performance gain. Specifically, the false negative rate decreases from 76.0% to 47.6% on average, corresponding to an improvement of 37.4% compared to the related pseudo label-based active learning method MORPH.

Förderung

  • Guangdong Basic and Applied Basic Research Foundation, China
  • Science and Technology Program of Guangzhou, China
  • Graduate Research and Innovation Program of Guangdong University of Foreign Studies
  • Guangdong Philosophy and Social Science Foundation
  • National Natural Science Foundation of China
  • Humanities and Social Sciences Planning Project of the Ministry of Education
  • Network and Data Security Key Laboratory of Sichuan Province

Zitieren

APA

Qiu, L., Liu, Z., Peng, B., Zhang, J., Liu, N., Wu, Z., & Wang, R. (2026). PLCDroid: enhancing android malware detection by mitigating pseudo-label noise in the presence of concept drift. The Computer Journal, 69(7), 1218-1236. https://doi.org/10.1093/comjnl/bxag021

Vancouver

Qiu L, Liu Z, Peng B, Zhang J, Liu N, Wu Z, et al. PLCDroid: enhancing android malware detection by mitigating pseudo-label noise in the presence of concept drift. The Computer Journal. 2026;69(7):1218-1236. doi:10.1093/comjnl/bxag021

BibTeX
@article{Qiu2026,
  author = {Qiu, Lingyu and Liu, Zhen and Peng, Bitao and Zhang, Jingjing and Liu, Ning and Wu, Zhihao and Wang, Ruoyu},
  title = {PLCDroid: enhancing android malware detection by mitigating pseudo-label noise in the presence of concept drift},
  journal = {The Computer Journal},
  year = {2026},
  volume = {69},
  number = {7},
  pages = {1218--1236},
  doi = {10.1093/comjnl/bxag021}
}
Kostenlos starten →