|
|
Using combined sequence and structural features to predict protein-ligand binding sites
Brabec, Matyáš ; Hoksza, David (advisor) ; Škoda, Petr (referee)
Protein interactions with various molecules are crucial for their function and, con- sequently, for the overall functioning of organisms. Studying these interactions is im- portant across many scientific fields, including medicine, where it is essential for drug development. Protein-ligand binding is a key type of these interactions, and a significant goal in bioinformatics is to develop reliable models for binding site prediction. The re- cent surge in the collection of protein structures, combined with the immense power of modern GPUs, has allowed the development of many machine-learning models. Notably protein language models, inspired by their counterparts in natural language processing, have been successfully applied throughout bioinformatics. In this thesis, we fine-tuned a protein language model for binding site prediction and sought to enhance its performance by incorporating various three-dimensional features of proteins. 1
|
|
|
Semi-automated data to ontology mapping
Hrubý, Kryštof ; Škoda, Petr (advisor) ; Nečaský, Martin (referee)
The most valuable data publication is in the form of linked data. The creation of linked data includes in particular the transformation of data into RDF and the appropriate use of identifiers and ontologies. However, the entire process is still challenging even for linked data experts. Therefore, the main goal of this thesis is to create a semi-automatic solution to facilitate the transformation of data into high-quality linked data. The basis of the proposed solution is the creation of a data model that the user can interactively edit manually or with the help of recommendations. These recommendations suggest model transformations based primarily on expert knowledge of data domains. The solution was implemented as a proof-of-concept web editor and is based on the visual modification of the schema of user provided data.
|
|
|
Development and deployment of SaaS solution for logistics automation in e-commerce
Půlpán, Michal ; Škoda, Petr (advisor) ; Zavoral, Filip (referee)
E-Commerce has seen rapid growth that significantly changed the retail landscape. This situation requires brands to do more than just offer products; they must present a distinct identity, maintain brand values, and establish deeper relationships with cus- tomers. Brands must focus on differentiating and drawing attention to themselves while keeping up with customer care. Traditionally, logistics communication during order de- livery is usually left to the shipping carrier, but it is an important element in the order lifecycle. It provides an opportunity to increase brand awareness and strengthen rela- tionship with the customer. This thesis proposes a solution to bridge the gap between streamlined shipping logistics and its use as a potential marketing channel. Introduces a SaaS platform tested in a real-world environment on thousands of parcels. Allowing sellers to manage parcel data transfers to carriers, print shipping labels, and efficiently track parcel statuses while providing shipment information to recipients through custom branded email notifications and tracking pages. This approach not only improves opera- tional efficiency, but also reinforces brand engagement throughout the delivery process. 1
|
|
|
Modular tool for creating games with artificial intelligence
Klika, Jan ; Škoda, Petr (referee) ; Kouřil, Jan (advisor)
This bachelor thesis describes the design and implementation of tool for facilitate the creation of artificial intelligence. The tool uses its own universal communication protocol, supports multiple types of games, is extensible for other games using dynamic libraries, supports the display of debugging information of artificial intelligence and is localized in English and Czech.
|
|
|
Scientific Web Portal Generator
Pekarovič, Peter ; Škoda, Petr (referee) ; Smrž, Pavel (advisor)
This thesis is concerned with implementation of scientific web portal and possibility of automatic generating of portal content. It describes possibilities of usage of alert services to acquire publications, analysis of keywords and categorization of publications. Thesis contains design and implementation of system for portal administration and portal itself.
|
|
|
Reinforcement Learning for RoboCup
Bočán, Hynek ; Škoda, Petr (referee) ; Smrž, Pavel (advisor)
Goal of this thesis is creation of artificial intelligence capable of controlling robotic soccer player simulated in SimSpark environment. Agent created is expanding capabilities of existing third party agent which provides set of basic skills such as localization on the field, dribbling with the ball and omnidirectional walk. Responsibility of the created agent is to pick the best action based current state of the game. This decision making was implemented using reinforcement learning and its method Q-learning. State of the game is transformed into 2D picture with several planes. This picture is then analyzed using deep convolution neural network implemented using C++ and DeepCL library.
|
|
|
Improving Bots Playing Starcraft II Game in PySC2 Environment
Krušina, Jan ; Škoda, Petr (referee) ; Smrž, Pavel (advisor)
The aim of this thesis is to create an automated system for playing a real-time strategy game Starcraft II. Learning from replays via supervised learning and reinforcement learning techniques are used for improving bot's behavior. The proposed system should be capable of playing the whole game utilizing PySC2 framework for machine learning. Performance of the bot is evaluated against the built-in scripted AI in the game.
|
|
|
Authorship Identification
Fabiánek, Ondřej ; Škoda, Petr (referee) ; Smrž, Pavel (advisor)
This bachelor's thesis deals with authorship identification based on knowledge of author's previous texts. The aim is to analyze existing methods of authorship attribution and create a system, which is capable of highly successful authorship identification. The system is based on a multivariate analysis and specializes at English books. Part of the solution is also a graphic user interface.
|
|
|
Automatic Keyword Suggestion
Šimara, Svatopluk ; Škoda, Petr (referee) ; Otrusina, Lubomír (advisor)
This thesis deals with the automatic keywords suggestion. The suggestion is based only on the statistic methods. For the analysis are used diploma thesis and similar documents. Statistic methods are detailed tested and evaluated by using these documents. For the final keywords suggestion ale chosen only the most successful methods. In the end, the suggested keywords are compared with the manual assigned keywords.
|
|
|
Quality Analysis of Electronic Dictionaries Transformation
Stehlíková, Petra ; Škoda, Petr (referee) ; Kouřil, Jan (advisor)
The bachelor's thesis deals with electronic dictionaries, their formats and quality analysis of their conversions. The thesis describes Lexical Markup Framework format in detail. It also discusses the capabilities of advanced algorithms such as LSA for conversion quality analysis and the tools that can be used for the analysis. Based on this theoretical knowledge the scripts in Python language were created to analyze dictionaries in Lexical Markup Framework format.
|
guest ::
RSS feed.