Year: 2025
Type: conference-paper
Abstract: The paper presents a research of parsing User Agent Strings (UAS) for machine learning and examines the features of relevant software libraries. A more accurate and structured method for identifying cyberattack characteristics is implemented without relying on complex regular expressions. The advantages of the proposed approach are highlighted, including improved data quality, reduced processing costs, and easier integration into machine learning (ML) pipelines. The article provides code examples and outlines implementation mechanisms for projects requiring the analysis of cyberattack data on client devices in financial institutions.
Authors: Dmytro Zavorotnii, Ivan Burlachenko
Institution: Petro Mohyla Black Sea National University
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4414957893
FWCI:
Cites:
Cited by:
Related to: