Vpliv dejavnikov na zaznavanje globokih ponaredkov na družbenih medijih

diplomsko delo

Kai Ličina (Author), Ines Kožuh (Mentor), Peter Čakš (Co-mentor)

Abstract

Diplomsko delo raziskuje vpliv dejavnikov na pravilnost ločevanja globokih ponaredkov od pristnih video posnetkov na družbenih medijih. Delo obravnava ključne koncepte, povezane z družbenimi mediji in širjenjem lažnih informacij, pri čemer poseben poudarek namenja tehnologiji globokih ponaredkov, ki omogoča ustvarjanje realističnih lažnih vsebin. Empirični del raziskave vključuje analizo vpliva uporabe družbenih medijev, vrste videoposnetkov, izobrazbe in predhodnega znanja o manipulaciji medijskih vsebin. Rezultati kažejo, da večina teh dejavnikov nima statistično pomembnega vpliva na pravilnost ločevanja globokih ponaredkov od pristnih video posnetkov, z izjemo Facebooka, in samozavesti pri prepoznavanju globokih ponaredkov, kjer smo pri zabavnih videoposnetkih zaznali statistično značilen vpliv. Rezultati te študije lahko služijo kot osnova za nadaljnje raziskave o zaznavi ponarejenih posnetkov med uporabniki družbenih medijev.

Keywords

globoki ponaredki;družbeni mediji;zaznavanje;

Data

Language:	Slovenian
Year of publishing:	2024
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[K. Ličina]
UDC:	070.16:077.5(043.2)
COBISS:	219784963
Views:	0
Downloads:	15
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Analysis of factors on perception of deepfakes on social media
Secondary abstract:	This thesis explores the impact of various factors on the accuracy of distinguishing deepfakes from authentic videos on social media. It addresses key concepts related to social media and the spread of misinformation, with a special focus on deepfake technology, which enables the creation of realistic fake content. The empirical part of the research analyses the influence of social media usage, types of videos, education, and prior knowledge of media manipulation. The results indicate that most of these factors do not have a statistically significant impact on the accuracy of distinguishing deepfakes from authentic video content, with the exception of Facebook and confidence in recognizing deepfakes, where a statistically significant effect was observed in the case of entertaining videos. The findings of this study can serve as a basis for further research on the detection of fake videos among social media users.
Secondary keywords:	deepfakes;social media;perception;diplomske naloge;
Type (COBISS):	Bachelor thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Medijske komunikacije
Pages:	1 spletni vir (1 datoteka PDF ([VIII], 47 f.))
ID:	24829070