Klasifikacija fragmenata datoteka nad otvorenim skupom razreda

master's thesis

• Title (english): Open-set recognition of file fragments
• Author: Laura Petan
• Mentor: Juraj Petrović (mentor)
• Committee member: Juraj Petrović (predsjednik povjerenstva)
• Committee member: Ana Sović Kržić (član povjerenstva)
• Committee member: Davor Petrinović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-04, Croatia

Abstract

Problem klasifikacije fragmenata datoteka jedan je od ključnih problema kojim se bavi računalna forenzika. Zbog složenosti problema koriste se tehnike strojnog učenja. U pravim forenzičkim slučajevima često nailazimo na razrede datoteka na kojima mreže nisu učene. U ovom radu proučavaju se načini klasifikacije na otvorenom skupu razreda. Opisana su i implementirana četiri algoritma. Algoritmi su uspoređeni s rezultatima na zatvorenom skupu podataka kako bi se ocijenila njihova uporabljivost.

Abstract (english)

The problem of file fragment classification is one of the key issues addressed by computer forensics. Due to the complexity of the problem, machine learning techniques are employed. In real forensic cases, we often encounter file classes on which the networks have not been trained. This paper studies methods for classification in an open set environment. Four algorithms are described and implemented. The algorithms are compared with results on a closed set to evaluate their usability.

• Keywords: fragmenti datoteka; neuronske mreže; klasifikacija; otvoren skup podataka; računalna forenzika
• Keywords (english): file fragments; neural networks; classification; openset; computer forensics
• Language: croatian
• URN:NBN: urn:nbn:hr:168:312108
• 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: Open access
• Created on: 2024-10-12 21:50:16