prikaz prve stranice dokumenta Primjena metoda strojnog učenja u domeni kibernetičke sigurnosti
No public access
master's thesis
Primjena metoda strojnog učenja u domeni kibernetičke sigurnosti
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. urn:nbn:hr:168:644554
Begušić, Domagoj
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Begušić, D. (2021). Primjena metoda strojnog učenja u domeni kibernetičke sigurnosti (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:644554

Begušić, Domagoj. "Primjena metoda strojnog učenja u domeni kibernetičke sigurnosti." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:644554

Begušić, Domagoj. "Primjena metoda strojnog učenja u domeni kibernetičke sigurnosti." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2021. https://urn.nsk.hr/urn:nbn:hr:168:644554

Begušić, D. (2021). 'Primjena metoda strojnog učenja u domeni kibernetičke sigurnosti', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 27 November 2024, https://urn.nsk.hr/urn:nbn:hr:168:644554

Begušić D. Primjena metoda strojnog učenja u domeni kibernetičke sigurnosti [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2021 [cited 2024 November 27] Available at: https://urn.nsk.hr/urn:nbn:hr:168:644554

D. Begušić, "Primjena metoda strojnog učenja u domeni kibernetičke sigurnosti", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:168:644554

Metadata
TitlePrimjena metoda strojnog učenja u domeni kibernetičke sigurnosti
Title (english)Applying Machine Learning Methods in the Domain of Cyber Security
AuthorDomagoj Begušić
MentorDamir Pintar (mentor)
Committee memberDamir Pintar (predsjednik povjerenstva)
Committee memberMihaela Vranić (član povjerenstva)
Committee memberLuka Humski (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2021-09-21, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Sustavi za obranu mreže od malicioznih paketa ili općenito pokušaja prodora rukovode s ogromnim količinama mrežnih podataka iz kojih se primjenom strojnog učenja može izvući znanje i prepoznati određene pravilnosti kako bi se obrana mreže automatizirala. Takvi podaci često nisu označeni ili ih je skupo označavati pa je potrebno pronaći model ili skupinu modela s obzirom na prirodu samog problema. Primjenom programskog okruženja Python i određenih modula za strojno učenje, obradu i vizualizaciju podataka razvijen je sustav za automatsko grupiranje, detekciju anomalija i klasifikaciju mrežnih podataka u svrhu odjeljivanja primjera koje mreža treba propustiti i onih koje ne smije. Objašnjena je cijela faza predobrade i eksploratorne analize podataka te svih korištenih modela. Mjere za vrednovanje modela opisane su i primijenjene na izrađene modele te su rezultati komentirani.
Abstract (english)
Network security systems while defending network from malicious packets or attempted intrusions manage huge amounts of network data from which knowledge and certain patterns can be recognized by applying machine learning methods in order to automate network defense. Such data are often unlabeled or expensive to label, so it is necessary to find a model or group of models given the nature of the problem itself. Using the Python software environment and certain modules for machine learning, data processing and visualization, a system for automatic grouping, anomaly detection and classification of network data has been developed in order to separate examples that need to be pass through network and those that should not be. The whole phase of pre-processing and exploratory data analysis and all used models are explained. Measures for model evaluation are described and applied to the developed models and finally the results are commented.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:644554
Study programmeTitle: Computing
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-13 09:45:24