Procjena domenske prenosivosti sustava za detekciju napada na mreži

master's thesis

• Title (english): Assessment of domain portability of network attack detection systems
• Author: Dora Pavelić
• Mentor: Mario Brčić (mentor)
• Committee member: Mario Brčić (predsjednik povjerenstva)
• Committee member: Mirjana Domazet-Lošo (član povjerenstva)
• Committee member: Nikica Hlupić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-05, Croatia

Abstract

Ovaj diplomski rad istražuje domensku prenosivost sustava za detekciju napada na mreži temeljenog na algoritmima umjetne inteligencije. Iako postoji veliki broj radova na temu algoritama za detekciju napada na mreži, samo mali dio se bavi problemom prenosivosti tih algoritama na druge domene. Ovaj rad se fokusira na istraživanje optimalnog procesa obrade podataka i primjenu različitih algoritama strojnog učenja kako bi se procijenila njihova učinkovitost na neviđenim skupovima podataka. Time se istražuje sposobnost modela da detektiraju nove napade, kao i različite varijacije već poznatih napada. Cilj je utvrditi u kojoj mjeri algoritmi umjetne inteligencije mogu održati visoku razinu performansi kada se suoče s novim i različitim prijetnjama izvan onih korištenih tijekom treniranja.

Abstract (english)

This thesis explores the domain transferability of network attack detection systems based on artificial intelligence algorithms. Although there is a large body of work on algorithms for network attack detection, only a small portion addresses the problem of transferring these algorithms to different domains. This study focuses on exploring the optimal data processing methods and the application of various machine learning algorithms to evaluate their effectiveness on unseen datasets. This approach investigates the models' ability to detect new attacks as well as different variations of previously known attacks. The aim is to determine the extent to which artificial intelligence algorithms can maintain high performance levels when confronted with new and diverse threats beyond those encountered during training.

• Keywords: domenska prenosivost; sustavi za detekciju napada; NIDS; algoritmi umjetne inteligencije; obrada podataka
• Keywords (english): domain transferability; intrusion detection systems; NIDS; artificial intelligence algorithms; data processing
• Language: croatian
• URN:NBN: urn:nbn:hr:168:746371
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2024-10-12 21:46:49