Classification of transactional anomalies in decentralized networks based on topological patterns for Deep graph learning
Radiotekhnika · 25.07.2026 · (225) · 51-56
DOI: 10.30837/rt.2026.2.225.03 ↗Kurz gefasst
Fachartikel in der Zeitschrift Radiotekhnika, online erschienen am 25.07.2026. Verfasst von V.V. Prosolov und 1 weiteren Autor:innen.
Abstract (Originalsprache)
The aim of the article is to develop and scientifically substantiate an approach to classifying transactional anomalies in decentralized blockchain networks based on stable topological patterns suitable for deep graph learning and transaction risk scoring. To achieve this aim, the following tasks are set: to analyse the limitations of existing approaches to the classification of blockchain threats; to identify topological patterns characteristic of targeted transactional attacks; to form an ML-oriented classification of anomalies; to describe the correspondence between categories of transactional anomalies, graph features, and the possibilities of applying deep graph learning models; and to determine practical directions for using the proposed classification in risk-scoring and AML/CFT monitoring systems. The article will be useful to specialists in information security who deal with data protection in distributed systems.
Zitieren
APA
Prosolov, V., & Khalimov, H. (2026). Classification of transactional anomalies in decentralized networks based on topological patterns for Deep graph learning. Radiotekhnika, 51-56. https://doi.org/10.30837/rt.2026.2.225.03
Vancouver
Prosolov V, Khalimov H. Classification of transactional anomalies in decentralized networks based on topological patterns for Deep graph learning. Radiotekhnika. 2026:51-56. doi:10.30837/rt.2026.2.225.03
BibTeX
@article{Prosolov2026,
author = {Prosolov, V.V. and Khalimov, H.Z.},
title = {Classification of transactional anomalies in decentralized networks based on topological patterns for Deep graph learning},
journal = {Radiotekhnika},
year = {2026},
number = {225},
pages = {51--56},
doi = {10.30837/rt.2026.2.225.03}
}
EUVIMED