Detekcija zloćudnog koda iz tokova podataka

master's thesis

• Title (english): Malicious code detection from data streams
• Author: Marko Mesarić
• Mentor: Marin Vuković (mentor)
• Committee member: Marin Vuković (predsjednik povjerenstva)
• Committee member: Miljenko Mikuc (član povjerenstva)
• Committee member: Dragan Jevtić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-08, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu analiziran je pojam zloćudnog programa i prikazana je podjela zloćudnog programa u karakteristične razrede. Analizirane su metode detekcije zloćudnog koda iz tokova podataka s naglaskom na metode koje detektiraju zloćudan kod iz podatkovnog dijela paketa mrežnog prometa. Dan je uvod u konvolucijske neuronske mreže. Prezentiran je ostvaren sustav za detekciju zloćudnog koda iz podatkovnog dijela TCP paketa temeljen na dubokom učenju, odnosno konvolucijskim neuronskim mrežama. Opisan je proces prikupljanja i pripreme skupa podataka korišten za učenje i validaciju ostvarenog modela. Prikazani su ostvareni rezultati i predložene su metode za buduća poboljšanja.

Abstract (english)

In this paper, the concept of malware is analyzed and the division of malware into characteristic classes is presented. Methods of detecting malicious code from data streams are analyzed, with emphasis on methods that detect malicious code from the network traffic payloads. An introduction to convolutional neural networks is presented. An implemented system for detecting malicious code from the TCP packet payloads based on deep learning, that is, convolutional neural networks, is presented. The process of data collection and preparation used for learning and validation of the realized model is described. The achieved results are presented and methods for future improvements are proposed.

• Keywords: zloćudni program; detekcija zloćudnog programa; klasifikacija mrežnog prometa; duboko učenje; duboke neuronske mreže; konvolucijske neuronske mreže; ids; tokovi podataka
• Keywords (english): malware; malware detection; network traffic classification; deep learning; deep neural networks; convolutional neural networks; ids; data streams
• Language: croatian
• URN:NBN: urn:nbn:hr:168:719990
• 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: 2023-01-13 10:25:20