{"id":35699,"date":"2025-12-19T12:21:55","date_gmt":"2025-12-19T08:21:55","guid":{"rendered":"https:\/\/cue.edu.ge\/?post_type=articles&p=35699"},"modified":"2025-12-19T12:23:42","modified_gmt":"2025-12-19T08:23:42","slug":"attack-detection-on-distributed-big-data-streams-using-machine-learning","status":"publish","type":"articles","link":"https:\/\/cue.edu.ge\/en\/articles\/attack-detection-on-distributed-big-data-streams-using-machine-learning\/","title":{"rendered":"Attack Detection on Distributed Big Data Streams Using Machine Learning"},"content":{"rendered":"

Gulnara Janelidze<\/strong><\/p>\n

Doctor of Engineering Sciences, Professor, Georgian Technical University<\/p>\n

janelidzegulnara08@gtu.ge<\/a><\/p>\n

\u00a0<\/strong><\/p>\n

Data Datashvili<\/strong><\/p>\n

PhD student,\u00a0 Samtskhe-Javakheti State University<\/p>\n

datadatashvili99@gmail.com<\/a><\/p>\n

\u00a0<\/strong><\/p>\n

Abstract<\/strong><\/p>\n

In the modern digital world, the volume of data is exponentially increasing day by day, creating the need to develop new security approaches. The use of distributed data systems, through which data flows move across different platforms, has become particularly relevant. However, this high degree of distribution and openness creates favorable conditions for cyber-attacks. It is worth noting that traditional intrusion detection tools perform best on relatively low-velocity data. They are ineffective in handling large-scale and high-velocity data, which cannot be adequately processed. This is why new methods must be adapted to handle big data in order to detect any signs of intrusion. In this regard, the use of machine learning methods is particularly important, as they can be applied both for anomaly detection and for identifying indicators of known attacks.<\/p>\n

This paper analyzes the importance of real-time detection of DDoS (Distributed Denial of Service) attacks on distributed big data and the associated challenges. It describes DoS attacks that directly target big data systems. Special attention is given to the timely detection of various types of attacks on distributed information flows using machine learning. To address this problem, a Random Forest model is proposed. An algorithm for detecting unauthorized intrusion into the system using a Random Forest classifier is developed, and a network intrusion detection model based on a deep parallel Random Forest is presented.<\/p>\n

The paper also analyzes the challenges associated with using Random Forest, which mainly concern data imbalance and computational complexity. Nevertheless, the Random Forest method retains its advantages for solving network anomaly detection tasks, particularly for real-time distributed data.<\/p>\n

Keywords:<\/em><\/strong> DDoS attacks, Random Forest method, deep parallel Random Forest.<\/p>\n

JEL<\/strong>: C55; C45; D83<\/p>\n

DOI: <\/strong>10.52244\/c2025.27<\/p>\n

The article is in Georgian.<\/strong><\/p>\n

References<\/strong><\/p>\n

Mark Talabis, Jason Martin, Robert McPherson, Inez Miyamoto, Information Security Analytics: Finding Security Insights, Patterns, and Anomalies in Big Data, ISBN- 978-0128002070, 2014.<\/p>\n

Pradip Kumar Das, Privacy and Security Issues in Big Data, ISBN 978-981-16-1006-6, 2021.<\/p>\n

Clarence Chio, David Freeman, Machine Learning and Security: Protecting Systems with Data and Algorithms, ISBN – 978-1491979907, 2018.<\/p>\n

Bojan Kolosnjaji, Huang Xiao, Peng Xu, Apostolis Zarras, Artificial Intelligence for Cybersecurity, ISBN – 978-1805124962, 2024.<\/p>\n

Ronny H\u00e4nsch, Handbook of Random Forests, ISBN \u2013 978\u2013981-322-405-6, 2025.<\/p>\n

\u00a0<\/strong><\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"_acf_changed":false},"gonisdziebebi":[103],"class_list":["post-35699","articles","type-articles","status-publish","hentry","gonisdziebebi-2025-en"],"acf":[],"_links":{"self":[{"href":"https:\/\/cue.edu.ge\/en\/wp-json\/wp\/v2\/articles\/35699","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cue.edu.ge\/en\/wp-json\/wp\/v2\/articles"}],"about":[{"href":"https:\/\/cue.edu.ge\/en\/wp-json\/wp\/v2\/types\/articles"}],"wp:attachment":[{"href":"https:\/\/cue.edu.ge\/en\/wp-json\/wp\/v2\/media?parent=35699"}],"wp:term":[{"taxonomy":"gonisdziebebi","embeddable":true,"href":"https:\/\/cue.edu.ge\/en\/wp-json\/wp\/v2\/gonisdziebebi?post=35699"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}