EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Methodology for the Detection of Contaminated Training Datasets for Machine Learning-Based Network Intrusion-Detection Systems

Joaquín Gaspar Medina-Arco iD, Roberto Magán-Carrión iD, Rafael Alejandro Rodríguez-Gómez iD, Pedro García-Teodoro iD

Sensors · 12.01.2024 · 24 (2) · 479

DOI: 10.3390/s24020479 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 12.01.2024. Verfasst von Joaquín Gaspar Medina-Arco und 3 weiteren Autor:innen. Beteiligte Einrichtung: Network Engineering & Security Group (NESG), University of Granada, 18012 Granada, Spain. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 69 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministerio de Ciencia e Innovación, University of Granada, MICIN/AEI/10.13039/501100011033.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the significant increase in cyber-attacks and attempts to gain unauthorised access to systems and information, Network Intrusion-Detection Systems (NIDSs) have become essential detection tools. Anomaly-based systems use machine learning techniques to distinguish between normal and anomalous traffic. They do this by using training datasets that have been previously gathered and labelled, allowing them to learn to detect anomalies in future data. However, such datasets can be accidentally or deliberately contaminated, compromising the performance of NIDS. This has been the case of the UGR’16 dataset, in which, during the labelling process, botnet-type attacks were not identified in the subset intended for training. This paper addresses the mislabelling problem of real network traffic datasets by introducing a novel methodology that (i) allows analysing the quality of a network traffic dataset by identifying possible hidden or unidentified anomalies and (ii) selects the ideal subset of data to optimise the performance of the anomaly detection model even in the presence of hidden attacks erroneously labelled as normal network traffic. To this end, a two-step process that makes incremental use of the training dataset is proposed. Experiments conducted on the contaminated UGR’16 dataset in conjunction with the state-of-the-art NIDS, Kitsune, conclude with the feasibility of the approach to reveal observations of hidden botnet-based attacks on this dataset.

Schlagwörter

Stichwörter der Autor:innen

NIDSanomaly detectionautoencodersdata qualitydeep learningmethodologyreal network datasets

Autor:innen und Institutionen

  • Network Engineering & Security Group (NESG), University of Granada, 18012 Granada, Spain.

ORCID-Profile: Joaquín Gaspar Medina-Arco, Roberto Magán-Carrión, Rafael Alejandro Rodríguez-Gómez, Pedro García-Teodoro

Förderung

  • Ministerio de Ciencia e Innovación
  • University of Granada
  • MICIN/AEI/10.13039/501100011033

Zitieren

APA

Medina-Arco, J. G., Magán-Carrión, R., Rodríguez-Gómez, R. A., & García-Teodoro, P. (2024). Methodology for the Detection of Contaminated Training Datasets for Machine Learning-Based Network Intrusion-Detection Systems. Sensors, 24(2), 479. https://doi.org/10.3390/s24020479

Vancouver

Medina-Arco JG, Magán-Carrión R, Rodríguez-Gómez RA, García-Teodoro P. Methodology for the Detection of Contaminated Training Datasets for Machine Learning-Based Network Intrusion-Detection Systems. Sensors. 2024;24(2):479. doi:10.3390/s24020479

BibTeX
@article{MedinaArco2024,
  author = {Medina-Arco, Joaquín Gaspar and Magán-Carrión, Roberto and Rodríguez-Gómez, Rafael Alejandro and García-Teodoro, Pedro},
  title = {Methodology for the Detection of Contaminated Training Datasets for Machine Learning-Based Network Intrusion-Detection Systems},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {2},
  pages = {479},
  doi = {10.3390/s24020479}
}
Kostenlos starten →