|
N-gram language model for a Czech spellchecker
Richter, Michal ; Bojar, Ondřej (referee) ; Straňák, Pavel (advisor)
The aim of this thesis is to explore the possibilities of using n-gram language models for spellchecking Czech texts and to implement an extension to the spellchecker which would be able to find such misspelled words that are true Czech words. Furthermore, the aim was to implement a simple web application which would present the extended spellchecker. The influence of using lemmatization and morphology analysis of words regarding the hit rate of finding misspelled words was also looked into. The methods of language modelling used in the thesis are described first. What follow, then, is the description of the procedure of the spellchecking program using language models. The next part shows the way of getting the data for language model training. In the following part, the evaluation of the language models created is presented. The final part shows the results achieved for each option of spellchecking.
|
|
The Typology of Decoration in the Books of Hoursfrom late Middle-Ages (Book of Hours from the National Museum in Prague, XVI.G.77)
Horáková, Dominika ; Kubík, Viktor (advisor) ; Royt, Jan (referee)
v angličtině/ English Summary The Typology of the Decoration of Books of Hours in the late Middle Ages (The Book of Hours from the Library of the National Museum in Prague, XVI G 77) The aim of the study is a basic introduction to typology problems in books of hours in the late Middle Ages; the text discusses liturgical, cultural-historical contexts, iconographical and typological variations in the decoration of books of hours. Chapter 4.2. is focused on one concrete example - the Book of hours from the Library of the National Museum in Prague (XVI G 77). This manuscript is very different when compared with examples from preceding chapter 4.1. This piece of common production is much more simple - not only in terms of textual variety, but also in terms of decoration (it lacks both a compact iconographical cycle and iconographical inovations or particularities). Therefore I try to create some extended descriptions of the decoration of the Book of hours from the Library of the National Museum in Prague, and also to develop my own hypothesis about the date and location of the origin of this manuscript. My hypothesis is based on the polemic with the three existing allusions about this manuscript and also on the proper iconographical and stylistic analysis itself. Finally, I propose that the full-page...
|
|
Application of Kalman Filtering
Svojík, Marek ; Anděl, Jiří (referee) ; Hlávka, Zdeněk (advisor)
The aim of this work is to discuss the use of the Kalman lter in some economical problems. Generally taken, the Kalman lter is a mathematical method (an algorithm) used for estimation of the non-observable component of a state. Especially, this approach will be applied to estimate the risk-neutral state price density of CALL options. In such case a non-linear relation between state and observed variables may be assumed, and the problem has to be linearized by Taylor expansion. In detail, the main Kalman ltering in the simple linear case will be presented in the rst chapter. In the second chapter, you can nd some application of that Kalman ltering in case of CALL options. The study of the extended Kalman lter and its application in case of a nonlinear state model and the use of the Taylor expansion can be found in Chapter 3. In the fourth chapter, we will be talking about estimating the risk-neutral price density of a CALL option. The corresponding outputs from the program R and the most important results of this work are summarized in the last Chapter 5.
|
|
Vektorová DIS cartridge pro Oracle
Nidrle, Václav ; Skopal, Tomáš (referee) ; Kopecký, Michal (advisor)
Oracle Text is an information retrieval system, which is included in Oracle database. It is built up on the basic Boolean model being further extended by features like assigning weights to individual tokens, calculating similarity of query and documents or fuzzy logic. The new database cartridge called Vector Text is implemented as the extension of Oracle Text, however it is based on the Vector space model. Thanks to this fact the possibilities of controlling the output amount and measuring the relevance of the returned documents are improved. For this purpose there is prepared a lot of functions for calculating token weight and similarity measure of query and document in Vector Text. The simple interface for their further extensions is also present there. The user interface of Vector Text cartridge kept the same principles used in the Oracle Text. Furthermore, the possibility of using special functionalities of Oracle Text being applicable in both models like thesaurus or stoplist have been preserved.
|
|
Procedure SDKL-Miner for data mining
Tegze, Miron ; Lín, Václav (referee) ; Rauch, Jan (advisor)
A new data mining GUHA procedure SDKL-Miner is presented which extends current data mining system LISp-Miner developed on the VŠE Praha. The procedure mines for various patterns based on evaluation of two two-dimensional contingency tables. A simple comparing quantifier is implemented. Tests and new results for this quantifier are presented together with set of examples of it's use. New directions of future development are proposed. The use of statistical dependency tests and the use of both vector and matrix majorization are analyzed in more details. The procedure is a result of continued development of the academic LISp-Miner system for knowledge discovery and data mining.
|
|
Volume data editor
Kavalík, Jiří ; Kolomazník, Jan (referee) ; Krajíček, Václav (advisor)
In the present work we want to design and implement simple tool for basic editing of volume data like resizing, rotating, tresholding retouching, which could be extended as needed. Currently, there are tools for volume data visualisation and simple editors for modding computer games with voxel-based graphics. There is not an editor with basic set of tools for editing volumes in resolution of medical scanners available to common user. The designed and implemented editor can run on common hardware and oth ers set of simple brushes and filters to the user and means to implement diferent tools as plugins for C# programmer.
|
|
Optimization distribution and collection of infusion solutions
Kravciv, Zbyněk ; Fábry, Jan (advisor) ; Pelikán, Jan (referee)
There are many distribution tasks, that vary in a number of vehicles, time windows, dividend or undivided delivery or if it is static problems or dynamic problems. In this essay I focus just on few of them. At first I put my mind to simple statistic distribution task with one vehicle. Later I extend it with time windows, when the point can be served by one car and by many cars. In the essay it will be solved the real task of distribution and delivery transportation of infusion in the hospitals. Because of the difficulty of solution I have to use the heuristic methods - Method of nearest neighbour, Savings method and Insert method. All these methods are modified by capacity requirements, time windows and also observence of the rules, which the drivers have to keep during a distribution. The aim is to minimize the distance travelled by the vehicles. And at least the company could be recommended the best solution.
|
|
Vector editor for creating animations
Kruliš, Martin ; Parízek, Pavel (referee) ; Zavoral, Filip (advisor)
The object of this work is to design and implement a vector editor with animation support. This program is designated for a very specialized group of users that want to create simple animations but find buying complex commercial tools unnecessary. The application will allow user to design basic vector objects, edit them and set them into motion using curved trajectories. Main features of the application are its modularity and flexibility, making it ready for future extending and customization.
|
|
Using Stratego/XT for Generation of Software Connectors
Malohlava, Michal ; Ježek, Pavel (referee) ; Bureš, Tomáš (advisor)
Software connectors are used in component based systems as a special entities modeling and realizing component interactions. Besides this behavior, connectors can provide extra functionality and benefits (e.g. logging, adaptation, monitoring). This approach requires generation of connector code with respect to requirements of components, a target environment and features specified at the design stage. In this thesis we show how to extend the existing connector generator [33] by the Stratego/XT transformation engine, which includes a language for implementing program transformations and a collection of transformation tools. We use the toolset to realize a simple method of defining connector implementation, which is use as a template for a process of generation source code.
|
|
Mathematical Programs for Dynamic Pricing - Demand Based Management
Hrabec, Dušan ; Karpíšek, Zdeněk (referee) ; Hoff, Arild (referee) ; Haugen, Kjetil Kare (advisor)
Tato disertační práce se zabývá vývojem, modelováním a analýzou poptávkově orientovaných úloh, které zahrnují marketingová, operační a logistická rozhodnutí. Úlohy jsou zvoleny tak, aby mohly být dále rozšířeny o koncept tzv. dynamického oceňování a jiných dynamických marketingových rozhodnutí. V práci jsou využity dvě základní poptávkově orientované úlohy: a) úloha kolportéra novin, která je zvolena pro její jednoduchou formu a která tak slouží jako nástroj pro ilustrativní ukázky rozhodovacích procesů v podobných typech úloh, a b) úloha návrhu dopravní sítě, kde jsou využity některé výsledky a znalosti získané při řešení úlohy kolportéra novin. Kolportér (či obecně maloobchodník) čelí náhodné poptávce, která může být postupně ovlivněna oceňováním, marketingovými (tj. reklamními) rozhodnutími a nakonec jejich kombinací. Poptávka obsahuje tedy náhodnou složku, která je pomocí přístupů stochastické optimalizace modelována ve specifickém tvaru (tj. aditivní či multiplikativní tvar). Závislost cena-poptávka je zachycena pomocí nelineární klesající poptávkové funkce, zatímco (vhodná) reklama vede ke zvýšení poptávky (běžně rostoucí s-křivka či konkávní funkce). Výsledky získané při řešení úlohy kolportéra novin s oceňováním jsou následně využity v úloze návrhu dopravní sítě. Tato stochastická úloha je modelována (reformulována) pomocí dvou přístupů stochastické optimalizace: wait-and-see přístup a here-and-now přístup. Jelikož tato implementace vede na lineární či nelineární celočíselnou (navíc scénářovou) úlohu, jsou v práci zmíněny taky výpočetní nástroje. Autor pro řešení používá (původní) tzv. hybridní algoritmus, což je kombinace heuristického (genetického) algoritmu a nástroje optimalizačního softwaru. Potenciální aplikace sestavených modelů, obzvláště v oblasti odpadového hospodářství, jsou diskutovány v závěrečné části disertační práce.
|
guest ::
