Sustav za detekciju zlonamjernih programa temeljen na konvolucijskim neuronskim mrežama

master's thesis

• Title (english): Convolutional Neural Network Based Malware Detection System
• Author: Juraj Prpić
• Mentor: Tomislav Hrkać (mentor)
• Committee member: Tomislav Hrkać (predsjednik povjerenstva)
• Committee member: Stjepan Groš (član povjerenstva)
• Committee member: Siniša Šegvić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-09-20, Croatia

Abstract

Kao posljedica sve veće digitalizacije svih sfera života, isplativost ucjene i krađe podataka raste. Paralelno tome upotreba zlonamjernih programa i pojava različitih varijanti su u uzlaznom trendu. Postoji više različitih pristupa problemu detekcije i klasifikacije zlonamjernih programa, neki od kojih uključuju strojno učenje. U radu je opisan sustav za detekciju zlonamjernih programa temeljen na konvolucijskim neuronskim mrežama. Kao polazišne točke koriste se prednaučeni modeli ResNet i EfficientNet. Prikazani su dobiveni rezultati te je provedena njihova analiza i usporedba.

Abstract (english)

As a consequence of the increasing rate of digitalisation of all our spheres of life, it is becoming more and more profitable to try to steal data and to attempt extortion. Alongside this, the usage of malware and the appearance of new variants show a rising trend. There are multiple different approaches to malware detection and classification problems, some of which are based on machine learning. In this thesis, a malware detection system based on convolutional neural networks is described. Pretrained models ResNet and EfficientNet are used as starting points. The obtained results are presented and an analysis and comparison are made between them.

• Keywords: zlonamjerni programi; konvolucijske neuronske mreže; detekcija; klasifikacija; prednaučeni modeli; ResNet; EfficientNet
• Keywords (english): malware; convolutional neural networks; detection; classification; pretrained models; ResNet; EfficientNet
• Language: croatian
• URN:NBN: urn:nbn:hr:168:870310
• 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 21:56:17