prikaz prve stranice dokumenta Sustav za otkrivanje upada u mrežu zasnovan na algoritmu strojnog učenja
  Download
PDF 2.69 MB
master's thesis
Sustav za otkrivanje upada u mrežu zasnovan na algoritmu strojnog učenja
2024. urn:nbn:hr:200:293219
Varvodić, Benjamin
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Communications
Chair of Electronics and Microelectronics

Cite this document

Varvodić, B. (2024). Sustav za otkrivanje upada u mrežu zasnovan na algoritmu strojnog učenja (Master's thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:293219

Varvodić, Benjamin. "Sustav za otkrivanje upada u mrežu zasnovan na algoritmu strojnog učenja." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2024. https://urn.nsk.hr/urn:nbn:hr:200:293219

Varvodić, Benjamin. "Sustav za otkrivanje upada u mrežu zasnovan na algoritmu strojnog učenja." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2024. https://urn.nsk.hr/urn:nbn:hr:200:293219

Varvodić, B. (2024). 'Sustav za otkrivanje upada u mrežu zasnovan na algoritmu strojnog učenja', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 16 February 2025, https://urn.nsk.hr/urn:nbn:hr:200:293219

Varvodić B. Sustav za otkrivanje upada u mrežu zasnovan na algoritmu strojnog učenja [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2024 [cited 2025 February 16] Available at: https://urn.nsk.hr/urn:nbn:hr:200:293219

B. Varvodić, "Sustav za otkrivanje upada u mrežu zasnovan na algoritmu strojnog učenja", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:200:293219

Please login to the repository to save this object to your list.
Metadata
TitleSustav za otkrivanje upada u mrežu zasnovan na algoritmu strojnog učenja
Title (english)Network intrusion detection system based on machine learning algorithm
AuthorBenjamin Varvodić
MentorMarijan Herceg (mentor)
MentorDenis Vranješ (sumentor)
Committee memberMario Vranješ (predsjednik povjerenstva)
Committee memberDenis Vranješ (član povjerenstva)
Committee memberRatko Grbić (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Communications)
(Chair of Electronics and Microelectronics)
Osijek
Defense date and country2024-10-01, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Telecommunications and Informatics
Abstract
U ovom radu istražena je primjena algoritama strojnog i dubokog učenja za detekciju upada u mrežu (engl. Intrusion Detection System, IDS). U radu je izrađena baza podataka simuliranih napada, pregled, odabir i implementacija različitih algoritama strojnog i dubokog učenja, te evaluacija njihove učinkovitosti za višeklasnu klasifikaciju. U bazi podataka realizirano je 5 vrsta napada: SYN flood, UDP flood, DNS amplification, smurf napad i port scanning. Predložena je struktura hibrida CNN-a (engl. Convolutional Neural Network) i LSTM-a (engl. Long short term memory) radi istraživanja uporabe dobrih strana oba algoritma. Algoritmi poput CNN-a i predloženog CNN LSTM hibrida, pokazali su visoku učinkovitost, dok je Histogram Gradient Boosting Classifier (HGBC) nadmašio ostale, ali pod cijenu potencijalnog pretreniranja. Zaključak rada je da je hibridni pristup CNN LSTM vrlo korisna kombinacija dvaju moćnih algoritama dubokog učenja, jedina prepreka je pronalazak optimalnih parametara.
Abstract (english)
In this paper, the application of machine and deep learning algorithms for an intrusion detection system is explored. The paper includes the creation of a simulated attack database, review, selection and implementation of various machine and deep learning algorithms as well as the evaluation of their multiclass classification effectiveness. 5 network attacks were succesfully recreated for use in the database: SYN flood, UDP flood, DNS amplification, smurf attack and port scanning. A hybrid convolutional neural network (CNN) and long short term memory (LSTM) structure is suggested, to use the good sides of both algorithms. Algorithms such as CNN and the suggested CNN LSTM hybrid have shown very high efficiency, while the Histogram Gradient Boosting Classifier (HGBC) surpassed all of the evaluated algorithms, albeit at the cost of potential overfitting issues. The conclusion of the paper is that the hybrid CNN LSTM approach is a very efficient combination of two exceptionally powerful deep learning algorithms, while the only obstruction being the tuning of parameters.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:293219
Study programmeTitle: University Study Automobile Computer and Communication
Study programme type: university
Study level: graduate
Academic / professional title: sveučilišni/a magistar/magistra inženjer/inženjerka automobilskog računarstva i komunikacija (sveučilišni/a magistar/magistra inženjer/inženjerka automobilskog računarstva i komunikacija)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2024-10-04 12:40:42