|
|
Perception of Generative Artificial Intelligence in selected Newsrooms of Domestic News Media with a Focus on Changes in Journalistic Ethics
Vaněček, Lukáš ; Moravec, Václav (advisor) ; Klimeš, David (referee)
The bachelor thesis delves into the subject of artificial intelligence and simultaneously explores its application in domestic editorial offices, along with the emerging ethical challenges associated with these technologies. In the theoretical section, in addition to introducing generative artificial intelligence itself and platforms such as ChatGPT or Midjourney operating on this basis, it describes the ethical domains in which the mentioned tools may instigate changes. The thesis includes a qualitative research component based on twelve in-depth semi- structured interviews with respondents from both public-service and private newsrooms actively utilizing generative artificial intelligence. Keywords Artificial Intelligence, AI, ChatGPT, Midjourney, ethics, transformation, machine learning, automated journalism, media Title Perception of Generative Artificial Intelligence in Selected Newsrooms of Domestic News Media with a Focus on Changes in Journalistic Ethics
|
|
|
Impact of European Central Bank and Federal Reserve System statements on cryptocurrency markets via sentiment analysis
Krejcar, Vilém ; Krištoufek, Ladislav (advisor) ; Čech, František (referee)
This study explores the impact of public statements from major central banks, specifically the FED and the ECB, on Bitcoin volatility from 2018 to 2021. Utilizing high-frequency data, we computed Bitcoin's volatility and extracted sentiment scores from the central banks' communications using two methods: the FinBERT language model and the state-of-the-art Generative AI GPT-4 model with tailored prompt. The GPT-4 model, capturing more nuanced senti- ment from text, was deemed superior. Our analysis involved comparing various models, with the HAR model emerging as the most e ective for this study. The research findings are particularly significant: negative sentiment from the ECB during the pandemic was associated with immediate and significant increases in Bitcoin volatility, indicating a market reaction of caution when faced with negative emission. These findings highlight the significant impact of central bank sentiment on Bitcoin volatility, confirming the initial hypothesis of this research. Additionally, the results provide a motivation to incorporate Genera- tive Artificial Intelligence into academic research as a tool for uncovering novel insights. JEL Classification C32, C55, C58, E58, G15 Keywords central banks, sentiment analysis, volatility, Bit- coin, GenAI, HAR, FED, ECB Title Impact of European...
|
|
|
Machine learning through geometric mechanics and thermodynamics
Šípka, Martin ; Pavelka, Michal (advisor) ; Monmarché, Pierre (referee) ; Maršálek, Ondřej (referee)
30. prosinec 2023 This thesis studies novel approaches to learning of physical models, incorporat- ing constraints and optimizing path dependent loss functions. Recent advances in deep learning and artificial intelligence are connected with established knowl- edge about dynamical and chemical systems, offering new synergies and improv- ing upon existing methodologies. We present significant contributions to sim- ulation techniques that utilize automatic differentiation to propagate through the dynamics, showing not only their promising use case but also formulating new theoretical results about the gradient behavior in long evolutions controlled by neural networks. All the tools are carefully tested and evaluated on exam- ples from physics and chemistry, thus proposing and promoting their further applications. 1
|
|
|
Comparative aAnalysis of Unsupervised Anomaly Detection Methods for Credit Card Fraud Detection
Jůzová, Anna ; Červinka, Michal (advisor) ; Janásek, Lukáš (referee)
In recent years, the increasing rate of cashless payments and online pur- chases has led to a rise in credit card fraud. Detecting fraudulent transactions poses a signifcant challenge for fnancial institutions, however, machine learn- ing has emerged as a promising tool. This thesis focuses on machine learning models for anomaly detection that have not received sufcient attention in pre- vious research. Specifcally, the study examines Isolation Forest, Local Outlier Factors, and One-Class Support Vector Machine. These models identify fraud- ulent payments as transactions that do not ft the learned pattern from past transactions. To optimise performance, the data are normalised using diferent normalisation techniques. Among the tested models, the Local Outlier Factor model with data normalised using the min-max method seems to be the most efective. JEL Classifcation C49, G21, K42 Keywords credit card fraud, machine learning, anomaly de- tection, data normalisation Title Comparative Analysis of Unsupervised Anomaly Detection Methods for Credit Card Fraud De- tection Author's e-mail anna.juzova11@gmail.com Supervisor's e-mail michal.cervinka@fsv.cuni.cz
|
|
|
Ab initio study of phase stability of multicomponent alloys
Fikar, Ondřej ; Brož, Pavel (referee) ; Černý, Miroslav (referee) ; Zelený, Martin (advisor)
Ab initio methods are based on purely theoretical findings of quantum physics that can be used to predict among others physical, chemical and mechanical properties of materials. Due to rapid increase in accessibility of computational resources in the recent decades the theoretical prediction of material properties became an integral part of materials design. This work is focused on theoretical prediction of phase stability and solubility of solid solutions. Ab initio calculations based on Density Functional Theory were performed using Projector-Augmented Waves method and thermal dependencies of thermodynamic quantities were obtained using phonon calculations and Monte Carlo simulations. Attention is paid to alloys mainly based on aluminium, silver and magnesium, which were investigated in order to assess the reliability and precision of theoretical predictions of solubility in the solid state. Phase stability of solid solutions was evaluated multiple times including different energy contributions and using various methods in order to determine the influence of each contribution and method on the prediction accuracy. Calculated solubilities are compared with experimental data provided using the CALPHAD method.
|
|
|
Detection of Harmfulness of Communication Partners and Their Networks
Kučera, Rostislav ; Homoliak, Ivan (referee) ; Očenášek, Pavel (advisor)
With the growing dependence of the population on electronic devices, the risk of data loss or misuse also increases. As the number of attacks in computer networks rises, systems for detecting malicious traffic become more important. The goal of this work is a theoretical analysis and implementation of modules for detecting malicious computer communication using machine learning methods, specifically a neural network model, and statistical analysis, which are deployed within the extended intrusion detection system Snort.
|
|
|
Information Extraction from Wikipedia
Jurišica, Rudolf ; Otrusina, Lubomír (referee) ; Smrž, Pavel (advisor)
The goal of this thesis is to reduce the number of unknown referenced entities in Czech Wikipedia articles. This has been achieved by using some existing solutions, created by the KNOT research group at FIT BUT, and then by creating a set of programs. These programs are automatically run every month, when a new version of Wikipedia is released. They will automatically add new names to the knowledge base, generate their derived forms, and edit the articles themselves directly on Wikipedia.
|
|
|
System for Recognizing Disinformation in Web Environment
Večerka, Lukáš ; Žádník, Martin (referee) ; Strnadel, Josef (advisor)
This work deals with the design, implementation, and verification of a system for automatic recognition of disinformation on the web. It addresses the issue of disinformation spread in the online environment and its impact on society. It focuses on training several Czech transformer language models for disinformation recognition and further automatic extraction of content from Czech online newspapers and their analysis using text classification and natural language processing through deep learning methods. The results of these analyses are then presented in a web user interface with the aim of providing a platform for verifying articles, authors, and sources. The interface could be used for data annotation by experts for continuous improvement of language models.
|
|
|
Detection of parking space availability based on video
Kužela, Miloslav ; Zelený, Ondřej (referee) ; Frýza, Tomáš (advisor)
Detekování obsazenosti parkovacích míst je často řešeno použitím senzorů umístěných v blízké lokaci parkovacího místa. Se vzrůstem strojového učení je možnost využití této technologie za použití kamer a detekčních algoritmů. Práce se zabývá právě vytvořením a použitím takového modelu k detekci obsazenosti parkoviště. Probírá existující modely a detektory, vytvoření vlastního datasetu s konkrétní strukturou, vytvoření a naučení různých typů modelů a probrání vysledků při testování daných modelů na vlastních záznamech z parkovací plochy. Poté následné vytvoření webové aplikace na které můžou návštěvnící parkoviště pozorovat obsazenost parkoviště. Vše za použití programovacího jazyka Python s knihovnami Torchvision.
|
|
|
Monitoring fitness activities using a wearable device
Toman, Adam ; Trčka, Tomáš (referee) ; Tofel, Pavel (advisor)
Bachelor thesis deals with the issue of fitness tracking with wearable device. Theoretical part describes the historical development of wearable devices, the base theory behind fitness activities and resistence training and methods of using of wearable devices for classification and evaluation of these activities. Aim for the practical part of this thesis was to develop an algorithm able to classify and evaluate activities through chosen recorded metrics. Practical part is followed by overall result evaluation and discussion.
|
guest ::
