|
|
Automatic Checking of Translation
Šimlovič, Juraj ; Kuboň, Vladislav (advisor) ; Dědek, Jan (referee)
Translation memories are becoming more and more popular with professional translators nowadays, especially in fields of software localization and translation of technical and official documents. Although commercial systems, which employ memory translation, provide some limited capabilities for automatic checking of translations, these are mostly of simple search-and-replace type. And none of these systems provide reasonable means of applying Czech morphology while checking. Professional translators could benefit from an automatic tool, which would provide more advanced rule-based checking capabilities, taking Czech and even English morphology into the process. Checking not only for correct use of terminology, but also for illicit translations and use of forbidden terms would be useful. This thesis investigates types of mistakes translators tend to make. Review of existing solutions for automatic translation checking for different languages is provided. An application is then suggested and developed, which attempts to search for some of the most frequent mistakes made in translations into Czech language, taking morphology into account while searching.
|
|
|
Linguistic Text Compression
Kazík, Ondřej ; Lánský, Jan (advisor) ; Dědek, Jan (referee)
The compression of texts written in natural language can exploit information about its linguistic character. It is shown that separation of coding of part-of-speech tags of a sentence (type of sentence) from the text alone can improve resulting compression ratio. For this purpose the tagging method NNTagger based on neural networks is designed. This thesis is focused on speci fication of the compression model of texts written in Czech. We propose methods of constructing of initial dictionaries and test their influence on the compression ratio.
|
|
|
C++ library for symbolic manipulation
Majdan, Ján ; Holan, Tomáš (advisor) ; Dědek, Jan (referee)
The aim of the bachelor thesis is the design and implementation of a library for symbolic manipulation with mathematical expressions, and the other intention of the work is the description and substantiation of used techniques and methods. The developed library should be easily expandable and modifiable. On the other hand, the library should be usable for long and complex computation. Demonstration of solutions for both of the mentioned requirements, that often require opposite approaches, is also part of the submitted work. The last, but not least, there is the topic that is concentrated on some hints and examples of using the library.
|
|
|
Market Research Web Portal
Dolejš, Ondřej ; Zavoral, Filip (advisor) ; Dědek, Jan (referee)
The presented work discourse about a creation of market research web portal. It follows the analysis of pivotal topics as well as the consequent portal design, after which it describes implementation details and attests resulting OnlinePanel portal performance. OnlinePanel is meant to offer effective and economical executing of market researches over the Internet. Onlinepanel offers flexible motivation system to respondents and respondent credibility control, arbitrary quota definition and unlimited respondent selection for research to administrator.
|
|
|
User anotation of web resources using microformats
Toušek, Jiří ; Vojtáš, Peter (advisor) ; Dědek, Jan (referee)
This thesis explores the feasibility of using the principles of user-assisted annotation for semantic annotation of web resources using microformats. The main concept explored is the approach using the user assistance not to mine the data but to create a set of "annotation rules" that could then be used repeatedly to annotate the target data, without requiring further user assistance. The thesis considers multiple ways to represent these annotation rules. An important attribute of this approach is a chance these annotation rules could withstand changes in the data annotated as well as minor changes in the web resource structure. A prototype implementation is created as a part of this thesis on which these concepts are then verified.
|
|
|
Process models of semantic web
Podzimek, Michal ; Zavoral, Filip (advisor) ; Dědek, Jan (referee)
The aim of this work is to explore the possibility of using the process models in the world of semantic web and the increase utilization of information from the semantic web. This software aims to add a lot of useful functionality to Trisolda, the infrastructure of the semantic web. It was analyzed that proposing of distributed system based on semantic services technology is an appropriate solution. The collaboration of several services is best described by the word "orchestration" that mean operating of executers by the conductor. BPEL (Business Process Execution Language) language is used for formalization of various processes. The proposed architecture is enlarged by pilot implementation that goal is to examine possible technologies that can be used to the realization. Several BPEL execution engines were tried before ActiveVOS platform was chosen as the most suitable one. The use of this platform is described in this work. The description of applied technologies and the confrontation with other existing projects is also a part of this thesis.
|
|
|
Graphical user interface for an RDF visualizer
Kóša, Peter ; Dokulil, Jiří (advisor) ; Dědek, Jan (referee)
In the present work we examine one implementation of a graphical user interface for a display and navigation in the RDF data. In addition, this single application is being briefly compared to several others, already existent alternative programs with the same purpose, followed by a short discussion of the application's overall pros and cons. The major focus of the thesis stays on the details of this particular implementation, evaluation of used algorithms and chosen components from C# and XAML technologies. Furthermore, we present suggestions for additional improvements and extensions that could be of use and therefore should be incorporated in the final product, should the application ever enter wholesale distribution.
|
|
|
Univerzální doporučovací systém
Cvengroš, Petr ; Vojtáš, Peter (advisor) ; Dědek, Jan (referee)
Recommender systems are programs that aim to present items like songs or books that are likely to be interesting for a user. These systems have become increasingly popular and are intensively studied by research groups all over the world. In web systems, like e-shops or community servers there are usually multiple data sources we can use for recommending, as user and item attributes, user-item rating or implicit feedback from user behaviour. In the thesis, we present a concept of a Universal Recommender System (Unresyst) that can use these data sources and is domain-independent at the same time. We propose how Unresyst can be used. From the contemporary methods of recommending, we choose a knowledge based algorithm combined with collaborative filtering as the most appropriate algorithm for Unresyst. We analyze data sources in various systems and generalize them to be domain-independent. We design the architecture of Unresyst, describe its interfaces and methods for processing the data sources. We adapt Unresyst to three real-world data sets, evaluate the recommendation accuracy results and compare them to a contemporary collaborative filtering recommender. The comparison shows that combining multiple data sources can improve the accuracy of collaborative filtering algorithms and can be used in systems where...
|
|
|
Analysis of existing Open Source alternatives to Microsoft Exchange Server, including verification of viability of the best chosen solution
DĚDEK, Jan
The subject of this thesis is open source alternatives to MS Exchange server. Microsoft Exchange Server is a system for cooperation, which enables management (storage, sharing ...) of corporate data. First of all emails, contacts and calendars, notes and tasks. All these data are synchronized and displayed on a user device (workstation, laptop, tablet, mobile phone ...). MS Outlook cares about well organised display of above mentioned data. This solution is very costly both in terms of software licenses, and in terms of hardware requirements. In this thesis will be presented and compared some open source groupware systems: a free alternative to MS Exchange Server. The best substitute will be selected and verified. The installation and configuration will be described on the virtual server platform Proxmox Virtual Environment. In the end the advantages and disadvantages of the mentioned alternatives will summarized. Notification of potential difficulties will be mentioned as well.
|
|
|
Semantic annotations
Dědek, Jan ; Vojtáš, Peter (advisor) ; Maynard, Diana (referee) ; Železný, Filip (referee)
Four relatively separate topics are presented in the thesis. Each topic represents one particular aspect of the Information Extraction discipline. The first two topics are focused on our information extraction methods based on deep language parsing. The first topic relates to how deep language parsing was used in our extraction method in combination with manually designed extraction rules. The second topic deals with a method for automated induction of extraction rules using Inductive Logic Programming. The third topic of the thesis combines information extraction with rule based reasoning. The core of our extraction method was experimentally reimplemented using semantic web technologies, which allows saving the extraction rules in so called shareable extraction ontologies that are not dependent on the original extraction tool. The last topic of the thesis deals with document classification and fuzzy logic. We are investigating the possibility of using information obtained by information extraction techniques to document classification. Our implementation of so called Fuzzy ILP Classifier was experimentally used for the purpose of document classification.
|
guest ::
RSS feed.