Primerjalna analiza orodij generativne umetne inteligence

Andrejek, Matej

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Primerjalna analiza orodij generativne umetne inteligence
Authors:	ID Andrejek, Matej (Author) ID Sternad Zabukovšek, Simona (Mentor) More about this mentor...
Files:	MAG_Andrejek_Matej_2026.pdf (2,39 MB) MD5: 2911CCF06019377219F437DCEC792396
Language:	Slovenian
Work type:	Master's thesis/paper
Typology:	2.09 - Master's Thesis
Organization:	EPF - Faculty of Business and Economics
Abstract:	Magistrska naloga obravnava področje generativne umetne inteligence (Gen AI) in primerja zmogljivosti izbranih orodij pri generiranju analitičnih odgovorov oziroma besedil. Namen raziskave je bil ugotoviti, ali med orodji ChatGPT, Microsoft Copilot in Google Gemini obstajajo razlike v kakovosti odgovorov na enake pozive ter ali plačljiva različica ChatGPT generira kakovostnejše rezultate kot njegova brezplačna različica. V teoretičnem delu naloge so predstavljeni temeljni koncepti umetne inteligence, strojnega učenja, globokega učenja in velikih jezikovnih modelov (LLM), ki predstavljajo osnovo sodobnih sistemov generativne umetne inteligence. Posebna pozornost je namenjena tudi opredelitvi in značilnostim generativne umetne inteligence ter razvoju, značilnostim in možnostim uporabe izbranih orodij. Empirični del naloge temelji na kvalitativni in kvantitativni primerjalni analizi štirih različic orodij generativne umetne inteligence: plačljive in brezplačne različice ChatGPT ter orodij Microsoft Copilot in Google Gemini. Analiza je bila izvedena na podlagi enotnega poziva, ki je zahteval pripravo analitičnega članka o razvoju španske nogometne lige LaLiga v obdobju sezon 2019/2020–2023/2024. Odgovori so bili ocenjeni glede na več meril, med drugim vsebinsko ustreznost, natančnost podatkov, analitično globino, strukturo besedila, kakovost virov ter prisotnost napak oziroma halucinacij. Rezultati raziskave kažejo, da med analiziranimi orodji obstajajo opazne razlike v kakovosti generiranih odgovorov. Najvišjo raven analitične poglobljenosti in metodološke doslednosti je dosegla plačljiva različica ChatGPT, sledil ji je Google Gemini, medtem ko sta brezplačna različica ChatGPT in Microsoft Copilot generirala bolj opisne in manj dosledne odgovore. Ugotovitve raziskave kažejo, da lahko izbira orodja in naročniškega modela pomembno vpliva na kakovost generiranih vsebin.
Keywords:	Umetna inteligenca, Generativna umetna inteligenca, Chat GPT, Google Gemini, Microsoft Copilot.
Place of publishing:	Maribor
Publisher:	M. Andrejek]
Year of publishing:	2026
PID:	20.500.12556/DKUM-98287
UDC:	004.8
COBISS.SI-ID:	289091843
Publication date in DKUM:	27.08.2026
Views:	195
Downloads:	28
Metadata:
Categories:	EPF
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:	CC BY-ND 4.0, Creative Commons Attribution-NoDerivatives 4.0 International

Link:	http://creativecommons.org/licenses/by-nd/4.0/
Description:	Under the NoDerivatives Creative Commons license one can take a work released under this license and re-distribute it, but it cannot be shared with others in adapted form, and credit must be provided to the author.
Licensing start date:	03.06.2026

Secondary language

Language:	English
Title:	Comparative analysis of generative artificial intelligence tools
Abstract:	The master's thesis addresses the field of generative artificial intelligence (Gen AI) and compares the performance of selected tools in generating analytical responses or texts. The aim of the research was to determine whether there are differences in the quality of responses to identical prompts among the tools ChatGPT, Microsoft Copilot, and Google Gemini, and whether the paid version of ChatGPT generates higher-quality results than its free version. The theoretical part of the thesis presents the fundamental concepts of artificial intelligence, machine learning, deep learning, and large language models (LLMs), which form the foundation of modern generative artificial intelligence systems. Special attention is also given to the definition and characteristics of generative artificial intelligence, as well as to the development, features, and potential applications of the selected tools. The empirical part of the thesis is based on a qualitative and quantitative comparative analysis of four versions of generative AI tools: the paid and free versions of ChatGPT, and the tools Microsoft Copilot and Google Gemini. The analysis was conducted using a unified prompt that required the preparation of an analytical article on the development of the Spanish football league La Liga during the seasons 2019/2020–2023/2024. The responses were evaluated according to several criteria, including content relevance, data accuracy, analytical depth, text structure, quality of sources, and the presence of errors or hallucinations. The results of the research indicate that there are notable differences in the quality of generated responses among the analyzed tools. The paid version of ChatGPT achieved the highest level of analytical depth and methodological consistency, followed by Google Gemini, while the free version of ChatGPT and Microsoft Copilot generated more descriptive and less consistent responses.
Keywords:	Artificial Inteligence, Generative Artificial Inteligence, Chat GPT, Google Gemini, Microsoft Copilot.

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back