Master’s theses

Master’s theses 177,038 records found  previous11 - 20nextend  jump to record: Search took 0.01 seconds. 
Two-phase scheduling with unknown speeds
Minařík, Josef ; Sgall, Jiří (advisor) ; Eberle, Franziska (referee)
Speed-robust scheduling is a two-stage scheduling problem with a makespan objective. We are given processing times of n jobs, number of machines m and number of bags b. We have to group the jobs into bags that are to be scheduled on machines of currently unknown speed. The goal is to minimize the worst-case ratio of our makespan and makespan of an adversary who does not have to create bags and assigns jobs directly to machines. So far, the problem has been mostly studied for b = m. We generalize previously known results for infinitesimal jobs (called sand) and prove that the best achievable competitive ratio is mb mb−(m−1)b . We present an algorithm for the case of identical jobs (called bricks) with competitive ratio at most 1.6 in the case b = m, improving the best previously known value of 1.8. We introduce a new category called p-pebbles, those are jobs with processing time at most p times the average load of a machine. Pebbles are half way between sand and the general case (called rocks). We present an algorithm for pebbles that has better robustness factor than the best known algorithm for rocks for small values of p (for p less than 2 − e e−1 in the case b = m). 1
CSV file validator according to the CSV on the Web W3C recommendations
Janda, Jan ; Klímek, Jakub (advisor) ; Svoboda, Martin (referee)
The comma-separated values (CSV) format is a popular format for tabular data on the web. It stores tables of data in a very simple textual way, but it stores only the values of the table without their meaning and structure. Users of a table must remove the ambiguity and guess the exact meaning of data. The standard called CSV on the Web provides recommendations for metadata about CSV tables on the web. The metadata can describe a CSV table, its structure, and the meaning of its values in the JSON-LD format. The standard allows a creator of a table to remove ambiguity, increase certainty, and create confidence. There are CSV tables together with their JSON-LD metadata descriptions, and each table should match its particular description. In this thesis, we develop a CSV file validator. The validator is a computer program which checks whether tables actually match their metadata descriptions. It reads a table with its description, compares them, and searches for discrepancies. This text describes the development process and its result.
Activating antitumor immune response using bispecific fusion proteins
Chytrá, Gabriela ; Vaněk, Ondřej (advisor) ; Černá, Věra (referee)
Natural killer (NK) cells are lymphocytes of the innate immune system that recognize and eliminate transformed and potentially harmful cells in a mechanism termed immunosurveillance. Malignant cells strive to escape immunosurveillance, and if successful, oncological disease develops. To restore immune recognition, immunotherapy utilizing NK cell-directed therapeutic fusion proteins can be employed. Therapeutic fusion proteins target tumour markers expressed on the surface of malignant cells and, at the same time, stimulate immune response through binding to NK cell activating receptors, for example receptor NKG2D or NKp30. A relevant example of a tumour marker is the HER2 receptor, which is often overexpressed in several types of cancer, most notably breast carcinoma. This thesis describes the preparation of several bispecific fusion proteins with potential use in immunotherapy. Bispecific fusion proteins consist of an NK cell activating ligand (ligand MICA or B7-H6) and nanobody targeting selected tumour marker (receptor HER2), which are connected by flexible glycine-serine linker. The constructs of fusion proteins were prepared in two configurations - with nanobody located on the N-terminus and the ligand on the C-terminus and vice versa. In addition, bispecific fusion proteins introducing...
Material picker: Material recognition in images using machine learning
Jurčák, Filip ; Vévoda, Petr (advisor) ; Šikudová, Elena (referee)
The process of setting material properties for realistic appearance after rendering is usually tiresome and often requires carefully crafted skill for fine-tuning the parameters, as different combinations of these parameters can produce different-looking materials. To simplify this process, we introduce a solution to the texture transfer problem by creating a pipeline containing several deep neural networks. These networks subsequently represent solutions to inverse rendering and material segmentation by predicting intrinsic scene characteristics, like diffuse and specular albedo, surface normals, glossiness, view vector, texture coordinates, and segmentation, all from a single image. Artists can subsequently plug these inferenced properties inside their 3D scene representations and thus reduce the time needed to iterate over several design ideas. To train these networks, we generated high-quality dataset of substantial size using physically-based techniques to ensure good generalization on real-world images. 1
At a Loss for Words: Capturing the Duality of Political Correctness (PC)
Djokićova, Eliška ; Parízek, Michal (advisor) ; Földes, Kristián (referee)
This thesis examines the concept of political correctness (PC). This social phenomenon has recently become a political tool used widely by right-wing politicians to describe the divide between the ruling "elites" and the "ordinary people." More generally, PC acts as an indicator of a broader cultural turn toward "post-material values" in the West. However, despite its prevalence, there is a lack of consensus on its definition and underlying purposes. This thesis wishes to contribute to the PC debate by 1) outlining the historical development of PC, 2) reviewing and synthesizing existing literature, 3) proposing a conceptualization of PC, and 4) developing a measurement tool for PC at the individual level. It is argued that PC has a dual nature: it is comprised of a normative (adherence to social norms and values) and a behavioral dimension (prescription of designated behaviors meant to be their expression). Moreover, it is hypothesized that a gap exists between these dimensions that can be influenced by reputational concerns (social desirability), perceived threats to freedom (psychological reactance), and political orientation. In this context, three pilot studies were conducted to validate the new measurement tools and examine the associations between PC's two dimensions and social desirability,...
6D pose estimation of objects in images
Cífka, Martin ; Šivic, Josef (advisor) ; Šikudová, Elena (referee)
The 6D pose estimation is an important computer vision task with applications in robotics, e.g. for manipulation or grasping, but also in computer graphics and augmented reality. Given an image, the task is to estimate the 3D rotation and 3D translation of the known object with respect to the camera. The task is even more challenging in an uncontrolled environment, e.g. when we do not have proper camera calibration. In that case, the focal length also needs to be estimated with the 6D pose. In this work, we address the issues of methods that work in such uncontrolled environments. First, we focus on FocalPose, a state-of-the-art method for joint estimation of object 6D pose and camera focal length. We review the method and propose several improve- ments. These include (i) re-deriving and improving the 6D pose and focal length update rule, (ii) replacing the model retrieval method, and (iii) changing the distribution of 6D poses and focal lengths used for synthetic training data rendering. These changes lead to improved results compared to the state-of-the-art FocalPose method. Second, to avoid often costly retraining of models for 6D pose estaimation, it is ben- eficial to consider methods with the ability to generalize to novel objects that have not been seen during training. These methods require a 2D...
Basics of computer operation - didactic materials for lower secondary school
Mazná, Michaela ; Brom, Cyril (advisor) ; Černochová, Miroslava (referee)
The Framework Educational Programme was revised in the Czech Republic in 2021 to include the basic principles of the Internet and computer networks in teaching computer science at ISCED level 2. According to the new revision, primary schools are supposed to start teaching at ISCED 2 from September 2024 at the latest, however, they face a lack of suitable teaching materials on this topic for this students' age group. Therefore, this work addresses this gap by creating four model lessons and materials for students and teachers of the lower secondary school at two different levels of difficulty (grades 6-7 and grades 8-9). The aim of the lessons is to explain the structure and basic principles of the Internet. The content of the lessons is based on known student preconceptions. The theoretical background of the work is based on the constructivist perspective "Knowledge in Pieces", ERR framework and the model of teaching by analogy. The lesson plans and corresponding materials for both levels of difficulty were created and tested through the combination of action and design-based research in six schools (13 classes, 258-271 students in total). The students' knowledge was examined by means of pre-post testing. The results indicate a significant positive effect of our lessons immediately after the lessons (d =...
Opportunities and Obstacles of Employment of Shelter Houses Clients
Kellnerová, Andrea ; Kopecký, Martin (advisor) ; Šerák, Michal (referee)
This master's thesis explores the opportunities and difficulties faced by clients of shelter houses in securing employment. The objective of the study is to identify factors influencing the job placement of this target group. The thesis introduces the institution of shelter houses, their legislative framework, and types of shelter houses from the perspective of target groups. It also delves into the phenomenon of unemployment, focusing on clients of shelter houses, its common causes, and its impacts on this group. The work highlights the employment opportunities available to shelter house clients, including options offered by active employment policies such as retraining, publicly beneficial work, and socially useful positions. It further discusses unemployment support, individual plans used in shelter houses, job activation, protected and open labor markets, training job positions, and also inappropriate employment opportunities. Finally, the thesis identifies the most common barriers to employment in the labor market, including low levels of education and qualifications, indebtedness, psychological and health disadvantages, employer prejudices, and limiting socio-cultural environments. The study also includes a quantitative survey conducted through questionnaires, aiming to identify factors...
Distance magic labelings
Pfeiffer, Hayden ; Gregor, Petr (advisor) ; Pangrác, Ondřej (referee)
Title: Distance Magic Labelings Author: Hayden Pfeiffer Department: Dept. of Theoretical Computer Science and Mathematical Logic Supervisor: doc. Mgr. Petr Gregor, Ph.D., KTIML, MFF UK Abstract: A distance magic labeling of a graph G is a bijection f : V (G) → {1, 2, . . . , |V (G)|} such that the sum of labels on the neighbourhood of each vertex is constant. A framework based on linear algebra has been developed using the notion of neighbour balance to determine whether there exists a distance magic labeling for a hypercube with dimension n. In this thesis, we extend this framework to all Cayley graphs on Zn 2 . We use this framework to reprove some known results from recent literature. We also use this framework to introduce the notion of component-wise distance magic labelings on Cayley graphs of Zn 2 . Keywords: distance magic labeling, Cayley graph, hypercube, neighbour balance iii

Master’s theses : 177,038 records found   previous11 - 20nextend  jump to record:
Interested in being notified about new results for this query?
Subscribe to the RSS feed.