|
|
Information Extraction from Loosely Structured Text
Minárik, Matej ; Bartík, Vladimír (referee) ; Burget, Radek (advisor)
Nowadays we are speaking about Web 2.0, which means the web of documents rather than the web of data. Documents are mostly unstructured, or just partially structured, but search engines need data in structured form in order to provide better search results. The process of extracting structured data from partially structured documents is the main goal of this work. In this work we are analyzing information extraction methods, namely classification methods, which need annotated training data, in order to create their inner model. We also analyze methods, which do not need training. These methods are initialized with a few data examples we are interested in extracting. We propose an extraction method in order to extract therapeutic indications and active substances from medical information sheets.
|
|
|
Evolutionary Design of Neural Networks
Beluský, Tomáš ; Vašíček, Zdeněk (referee) ; Minařík, Miloš (advisor)
The work deals with the development of the genetic algorithm, which designs the structure and learning of the neural networks. The fitness function also includes the number of hidden neurons, and thus we obtain the most optimal structure, which is reachable. The own versions of the operators are presented, which manage the entire process of evolution. The result of the work is a library for evolutionary design of neural networks. Moreover, graphical interface for setting parameters and displaying the results was created. In the experimental part the design is compared with other systems and algorithms. Finally, results are reviewed and the process for the following development of the system is outlined.
|
|
|
Online Flash Communication Platform
Krupa, Peter ; Minařík, Miloš (referee) ; Vašíček, Zdeněk (advisor)
This thesis dealt with issue of creating communication platform via Flash technology. It includes study of the possibilities this technology brings along with cooperation of using other available open-source sources for designing communication software. Beside this analysis, my work contains guideline, how to proceed in a matter of designing this kind of a project and working plan I've chosen. I've done my best to put to my work all information about this problem and tried to show a possible way, how to finish the task as this project was.
|
|
|
Recognition of Handwritten Mathematical Expressions by Neural Network
Šimeček, Josef ; Bidlo, Michal (referee) ; Minařík, Miloš (advisor)
This thesis deals with recognition of handwritten mathematical expressions by neural network. Neural network, mathematical syntax, issues of handwritten mathematical expressions and existing mathematical expressions recoginition systems are described in theoretical part. Practical part includes my own proposal of recognition system. Each part (step) of the system is analyzed in detail. Next part of thesis contains tests performed by implemented recognition system. The results are discussed and reviewed.
|
|
|
Genetic Programming with Brainfuck Language
Minařík, Miloš ; Bidlo, Michal (referee) ; Gajda, Zbyšek (advisor)
This thesis deals with the research of possible usage of brainfuck language for genetic programming. In the theoretical part, there are described basic mechanisms of genetic programming (selection, crossover, mutation), their different versions and usage. Practical part includes chapters considering implementation of program interface for this thesis and the experiments done. Results from the experiments are discused in the final chapter, where the future development in the are of genetic programming with brainfuck language is also mentoined.
|
|
|
Detection of Elliott Waves Using Neural Networks
Grega, Martin ; Vašíček, Zdeněk (referee) ; Minařík, Miloš (advisor)
This thesis deals with application of artificial neural networks for detection of Elliott waves in static time-series. It is focused mainly on use of simple committee machine consisting of multilayer perceptrons trained by resillient propagation algorithm. Thesis contains design and implementation of application for detection of impulsive waves on input signal.
|
|
|
Tools for Automatic Creation of Presentations
Uherčík, Maroš ; Minařík, Miloš (referee) ; Kadlček, Filip (advisor)
This bachelor thesis deals with design and implementation of tool for automatic presentation creation in OOXML format. Work contains theoretic analysis of the OOXML format, focusing on presentation, analysis and design of tool and implementation description. Practical part of this work is implementation of designed tool. The final part shows obtained results, demonstration of tool usage and proposal of future works.
|
|
|
FITkit Music Synthesizer
Melichar, Vojtěch ; Minařík, Miloš (referee) ; Vašíček, Zdeněk (advisor)
This bachelor's thesis deals with sound synthesis methods and with design of a synthesizer for FITkit platform. In the first part of the thesis, there are described methods of sound synthesis and the history of sound synthesis is also stated here. Next part contains a brief characteristic of sound chip SID 6581. In the next part of the thesis, there is a design of each part of the synthesizer: oscillator, envelope generator, voice and filter. Then, these components are put together to make up an synthesizer. Correctness of the implementation is verified by a simulation in ModelSim 6.6d. Individual components are simulated separately and then in mayor groups.
|
|
| |
|
|
Self-Modifying Programs in Cartesian Genetic Programming
Minařík, Miloš ; Slaný, Karel (referee) ; Sekanina, Lukáš (advisor)
During the last years cartesian genetic programming proved to be a very perspective area of the evolutionary computing. However it has its limitations, which make its use in area of large and generic problems impossible. These limitations can be eliminated using the recent method allowing self-modification of programs in cartesian genetic programming. The purpose of this thesis is to review the development in this area done so far. Next objective is to design own solutions for solving various problems that are hardly solvable using the ordinary cartesian genetic programming. One of the problems to be considered is generating the terms of various Taylor series. Due to the fact that the solution to this problem requires generalisation, the goal is to prove that the self-modifying cartesian genetic programming scores better than classic one for this problem. Another discussed problem is using the self-modifying genetic programming for developing arbitrarily large sorting networks. In this case, the objective is to prove that self-modification brings new features to the cartesian genetic programming allowing the development of arbitrarily sized designs.
|
guest ::
