EUVIMEDEUVIMED entdecken
Crossref

METHOD FOR FAST ANOMALY DETECTION IN IT SYSTEMS LOG RECORDS

Glib Pakharenko, Oksana Shpak, Tetiana Skolozdra, Volodymyr Samotyy

Measuring Equipment and Metrology · 2026 · 87 (1) · 63-69

DOI: 10.23939/istcmtm2026.01.063 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Measuring Equipment and Metrology, 2026. Verfasst von Glib Pakharenko und 3 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To detect targeted cyberattacks in a timely manner, security analysts are forced to filter out numerous false positives in antivirus logs. The values of some categories of file attributes have a probability distribution with a “heavy tail”, and malware has a “heavier” tail than common programs. The article proposes a method for detecting events in antivirus logs that are real malware events, and which is based on the peculiarities of the probability distribution of malicious and legitimate programs. A theoretical justification is made for the assumption of an exponential distribution of antivirus event attribute values. The essence of the approach is to sequentially select the most probable set of antivirus records with the least probable values of certain attributes. This allows us to obtain in the end a subset of the event log that is convenient for manual processing by the analyst with a much higher concentration of events of malicious programs of directed cyber-attacks, which are very rare, in other words, anomalous. The effectiveness of the proposed method is experimentally confirmed, recommendations are provided regarding the characteristics of event attributes and the order in which to filter the antivirus log, the number of attributes after which filtering should not be performed, and the probability limits of the subset of events selected at each iteration of the proposed method.

Zitieren

APA

Pakharenko, G., Shpak, O., Skolozdra, T., & Samotyy, V. (2026). METHOD FOR FAST ANOMALY DETECTION IN IT SYSTEMS LOG RECORDS. Measuring Equipment and Metrology, 87(1), 63-69. https://doi.org/10.23939/istcmtm2026.01.063

Vancouver

Pakharenko G, Shpak O, Skolozdra T, Samotyy V. METHOD FOR FAST ANOMALY DETECTION IN IT SYSTEMS LOG RECORDS. Measuring Equipment and Metrology. 2026;87(1):63-69. doi:10.23939/istcmtm2026.01.063

BibTeX
@article{Pakharenko2026,
  author = {Pakharenko, Glib and Shpak, Oksana and Skolozdra, Tetiana and Samotyy, Volodymyr},
  title = {METHOD FOR FAST ANOMALY DETECTION IN IT SYSTEMS LOG RECORDS},
  journal = {Measuring Equipment and Metrology},
  year = {2026},
  volume = {87},
  number = {1},
  pages = {63--69},
  doi = {10.23939/istcmtm2026.01.063}
}
Kostenlos starten →