National Repository of Grey Literature
Not a member yet
574846 research outputs found
Sort by
Methods and procedures of feedback as a motivational factor in further education of pedagogical staff
No full textYour thesis focuses on analyzing specific feedback delivery procedures and their utilization to motivate educational staff for further professional development. It describes various forms and methods of feedback provided by supervisors to educators. The impact of feedback on educational staff and whether it enhances work productivity and fosters interest in further professional development programs (DVPP) is examined. The thesis also monitors the established DVPP system in specific state-run preschools KEYWORDS Feedback, employee motivation, further education of pedagogical staff, organization in preschools and primary schools
Proposition for implementation of the CI system to the firm
No full textThis bachelor’s thesis addresses the topic of competitive intelligence in general, followed by specific requirements of a particular company that is seeking to implement a process of CI into its workings. Based on the requirements, appropriate methods and procedures will be selected and compiled into concrete and detailed instructions, enabling the company to establish and manage its own competitive intelligence process
Marketing Mix of the Company
No full textThis bachelor thesis focuses on the marketing mix of the restaurant chain Faency Fries s.r.o. The thesis is divided into three initial parts. The first, theoretical part discusses the key concepts of marketing, including the marketing and communication mix. The second, analytical part focuses on the current situation of the company using in-depth analyses. The third and final part is aimed at proposing specific recommendations and measures, based on the data obtained, to improve the company's marketing and communication mix and to contribute to the improvement of the company's marketing mix
Improvements to Mobile Device Movement Simulation in JShelter
No full textModerní chytré telefony vybavené různými senzory, jako jsou akcelerometry a gyroskopy, jsou náchylné ke zneužití a umožňují útočníkům sledovat, odposlouchávat nebo vytvářet digitální otisky bez jejich souhlasu. JShelter proti těmto hrozbám bojuje tím, že data ze senzorů maskuje pomocí realistických simulací. Tato práce představuje využití neuronové sítě k simulaci věrohodných pohybů uživatele, čímž překonává omezení modelů založených na fyzice. Její implementace je snadno zasazena do webového rozšíření JShelter a demonstruje vylepšené obfuskační techniky, které zvyšují soukromí uživatelů, aniž by výrazně omezily funkčnost rozšíření. Vyhodnocení pomocí nejmodernějšího systému rozpoznávání aktivit ukazuje, že ve více než 98 % případů klasifikátor správně identifikuje aktivitu simulovanou systémem JShelter, což dokazuje realističnost a robustnost systému generování vzorků
Sales Promotion and Advertising of Selected Company
No full textThis bachelor thesis analyzes the marketing communication of the bershka brand in the Czech Republic, focusing on its retail store in Brno. It examines the communication mix, competition and survey results. Based on the analysis, specific improvements are proposed in the areas of point-of-sale design, influencer collaboration and online communication. The goal is to increase customer loyalty, improve store orientation and strengthen brand image on the market
Identifying Narrative Similarity
No full textComparing story similarity presents a complex challenge. This diploma thesis addresses this problem by designing and implementing a system that utilizes large language models for aspect-oriented story analysis – focusing on plot, characters, setting, and themes – to enable granular and interpretable comparisons. A significant contribution of the thesis is a web user interface that allows for exploratory search, visualization of similarities across aspects, and detailed examination of analyses. The system also integrates an agent based on Retrieval-Augmented Generation (RAG) principles for story discovery. For the evaluation of the system and the contribution of large language model analyses, a new dataset was created for evaluating the retrieval of similar stories. Experimental results confirm that data from the multi-aspect analysis, integrated into the search process, have a positive impact on the observed metrics. The thesis thus demonstrates a functional system and confirms the benefit of granular, aspect-based analysis, presented through a user-oriented interface, for exploring similarity between stories
Proposal to improve the marketing mix of the fitness center
No full textTato bakalářská práce analyzuje a navrhuje zlepšení marketingového mixu fitness centra AZ Fitness v Brně s cílem zvýšit návštěvnost o víkendech. Na základě modelu 4P a vytvořených zákaznických person byla provedena strukturovaná pozorování a rozhovory s majitelem a zaměstnanci. Součástí je také srovnání s dvěma konkurenčními fitness centry. Výsledky ukazují klíčové oblasti ke zlepšení v oblasti služeb, cen, viditelnosti a propagace. Navržené změny mají za cíl pomoci AZ Fitness přilákat více klientů a posílit jeho postavení na místním trhu
Location Computation Based on Images from Optometrical Systems
No full textThis thesis addresses the problem of determining the position of laser beam sources based on their imaging in open space. It builds upon previous work focused on capturing and localizing the beam within an image. As part of this thesis, an implementation of a camera orientation calibration algorithm is developed, based on imaging beams from sources with known positions. Furthermore, the thesis focuses on optimizing an existing method for determining the position of light sources using 2 to 4 cameras. Finally, it explores possible approaches for estimating source positions from images captured by a single camera and filtering these positions using additional sources of information
Using Reinforcement Learning for Vehicle Autonomous Driving in CARLA Simulator
No full textThis thesis focuses on researching the possibilities of autonomous driving in the CARLA simulator. It summarizes the key ideas and approaches that lead to the successful use of neural networks for vehicle control. It presents an overview of the most effective algorithms and methods that comprise modern solutions. The aim of the paper is to design and implement a modular system for autonomous driving based on reinforcement learning. A system architecture is defined, describing the roles of each module and how they communicate with each other. Based on the proposed architecture, the individual modules have been implemented independently. For some modules, different approaches are compared, while others are evaluated in a single implementation. The main aim is to simplify the perception of the surrounding world as much as possible in order to use reinforcement learning algorithms, which have not been sufficiently explored in the CARLA simulator environment
Usage of Neural Networks in Bitcoin Network Analysis
No full textPseudonymity and the absence of a central control entity in the Bitcoin cryptocurrency attract the attention of malicious actors seeking to exploit blockchain technology for money laundering, terrorism financing, fraud, and the purchase of illegal goods on darknet marketplaces. The identification of fraudulent activities and illegal purchases is a key task for security forces, while the detection of money laundering and terrorism financing is a primary focus for institutions combating financial crime. This thesis focuses on the use of deep learning for forensic analysis of the Bitcoin network blockchain. It outlines the main challenges that need to be addressed for machine learning methods to be effectively applied in forensic practice, and analyzes the content of the most commonly used anonymized dataset employed to train models for identifying transactions associated with illegal activity. Through reverse engineering, I was able to deanonymize 47 % of the original labels, which enabled the identification of 100 % of the transactions contained in this dataset. This thesis presents an open dataset that corrects critical flaws in the original anonymized dataset, thereby providing a necessary foundation for the deployment of deep learning methods in practical forensic applications. Finally, the thesis experimentally evaluates and compares traditional machine learning techniques with state-of-the-art deep learning methods based on graph neural networks, using both the original anonymized dataset and the newly introduced open dataset