Institutional Repository of Vilnius University
Mašininiu mokymusi paremtas anomalijų aptikimas /
Abstract
dc:descriptionThe aim of the bachelor thesis is to determine whether the new hybrid algorithms OPTICS – DBSCAN, OPTICS – K-means, OKSCAN can more accurately detect anomalies and make fewer errors with simulated DDoS attack data than the OPTICS algorithm. The work describes what DDoS attacks are and their principle of operation and what hybrid algorithms are. This information is used later to determine the methodology of the experiments, which will be used the same for all three algorithms, so that the test conditions are fair and the same for all hybrid algorithms. The principle of operation of each algorithm, the obtained results, advantages and disadvantages are also described. Finally, the work results show that the OPTICS – DBSCAN algorithm improves the performance of the OPTICS algorithm the most, the OKSCAN algorithm also improves the performance of the OPTICS algorithm, but not as much as OPTICS – DBSCAN, and the OPTICS – K-means algorithm does not improve the performance of the OPTICS algorithm. When choosing whether to use one of these algorithms instead of the OPTICS algorithm, it is worth considering the advantages and disadvantages of each and how they perform not only with simulated data, but also with real DDoS attack data.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Martinonis, Justinas,
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:ELABAETD210646650&prefLang=en_US