Uporaba umetne inteligence na področju varnosti gesel

diplomsko delo

Matej Malek (Author), Iztok Fister (Mentor), Viktor Taneski (Co-mentor)

Abstract

V diplomski nalogi smo predstavili osnove umetne inteligence oziroma strojnega učenja in njeno uporabo v varnosti sodobnih informacijskih tehnologij ter rešitev. Podali smo razloge, zakaj je njena uporaba koristna in potrebna ter kako se z njo že srečujemo v vsakdanjem življenju. Navedli smo tudi bistvene težave in napade, s katerimi se srečujemo, ter kako jih ljudje rešujejo s pomočjo umetne inteligence. V teoretičnem delu smo se osredotočili na razjasnitev pojmov, zgodovino napadov in razlago uporabljenih algoritmov. V praktičnem delu pa smo opravili primerjavo treh algoritmov strojnega učenja za ugotavljanje moči gesla, s čimer smo prišli do zaključka, da so nevronske mreže, čeprav najbolj časovno potratne, tudi najboljša izbira.

Keywords

K-najbližji sosed;klasifikacija;nevronske mreže;odločitvena drevesa;strojno učenje;diplomske naloge;

Data

Language:	Slovenian
Year of publishing:	2024
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[M. Malek]
UDC:	004.8:004.056.523(043.2)
COBISS:	220418819
Views:	0
Downloads:	20
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	The use of artificial intelligence in the field of password safety
Secondary abstract:	In the thesis, we presented the fundamentals of artificial intelligence, or machine learning, and its application in the security of modern information technologies and solutions. We provided reasons why their use is beneficial and necessary, as well as how we encounter them in everyday life. We also outlined essential problems and attacks we face and how artificial intelligence addresses them. In the theoretical part, we focused on clarifying concepts, the history of attacks, and explaining the algorithms used. In the practical part, we conducted a comparison of three machine learning algorithms for assessing password strength, where the verdict of the experiment was, that although neural networks needed the most time for model training, they proved to be the most efficient and accurate.
Secondary keywords:	K-nearest neighbors;classification;neural networks;decision tree;machine learning;bachelor's degrees;
Type (COBISS):	Bachelor thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Informatika in tehnologije komuniciranja
Pages:	1 spletni vir (1 datoteka PDF (XI, 54 f.))
ID:	24590262