Institutional Repository of Vilnius University
Mašininiu mokymu grįstų apsaugos metodų prieš DDoS atakas analizė /
Abstract
dc:descriptionIn this paper, machine learning for protection against DDoS attacks is investigated. Damage caused by DDoS attacks together with the number and accessibility of them is increasing. Tra‑ ditional methods of protection against DDoS can decrease these numbers, however, the attacks still remain an immense problem for software. Because of this, it is beneficial to use a modern technology such as machine learning to solve this problem. Machine learning’s biggest advantage is its ability to recognize DDoS attacks by using patterns learned from an immense number of DDoS attacks and normal traffic examples, rather than from human‑defined rules. Machine le‑ arning models can differ significantly from one another, thus, it is important to test and compare them. Therefore, in this paper models are tested by creating and inputting traffic data collected during a DDoS attack. The findings show which model performed better, how models’ accuracy depended on the speed at which attack requests were sent and indicate which variables influence the accuracy of a specific model the most.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Čepulis, Rytis,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD192846213&prefLang=en_US