|
|
The Exploitation of Linguistic Information in EBMT
Týnovský, Miroslav ; Kuboň, Vladislav (advisor) ; Žabokrtský, Zdeněk (referee)
Example-based machine translation (EBMT) is a corpus-driven method of machine translation. It builds the translation using analogy of the input text with a translation already made. The benefit of using linguistic knowledge within EBMT is the subject of this thesis. Two language pairs are covered: Czech-English and Czech-German. The thesis covers gathering annotated parallel Czech-German data, design and implementation process of an experimental EBMT system, and the effort to improve it using linguistic knowledge. Detailed evaluation and comparison of both the baseline EBMT and the linguistically enhanced system are described. Evaluation has been done using machine and human evaluation methods. The three automatic evaluation methods are BLEU, NIST and METEOR. The linguistic enhancement of the baseline EBMT system includes comparisons of the input sentence with the examples in the translation memory based on morphology and syntax.
|
|
|
A tool for generating code supporting reading of text files
Černý, Ondřej ; Kolovratník, David (advisor) ; Týnovský, Miroslav (referee)
In the present work we deal with the concept and implementation of tool AutoComp for generating programs for processing structured text files. The tool accepts description of the text whith making use of regular expressions and state machines. The output of the tool is a code in a programming language which gives interface for integrating code to a final program. It is possible to define points for calling specified functions and returning controll to the caller program in the input description. The use of generated code is to tokenize text or parse elementary gramatics. The tool provides two target programming languages and few character encodings. The concept was focused on extensibility to more languages and character encodings.
|
|
|
Algorithm for synchronization of radar air situation
Vilušínský, Zdeněk ; Kolovratník, David (advisor) ; Týnovský, Miroslav (referee)
The subject of this work is analysis and creation of an algorithm which would be able to find, based on input data from reference and measured radar air situation, a transformation od measured situation towards reference and finally valuate its usefulnes in practice. Program input is a file with measured data from both air situations. This work includes several examples of input date, which were used for testing purposes of algorithm. Output are values of several errors that manifests in examined situation.
|
|
|
Learning management system
Kinšt, Jakub ; Týnovský, Miroslav (advisor) ; Krůza, Oldřich (referee)
Presented bachelor thesis realizes web application helping lecturers and students of an arbitrary full-time presentation course. It provides simple, yet effective communication between the lecturer and his students, and among the students themselves via comments, private messages or discussion forum. It offers possibility to publish students' results, to test students via online assignments and to provide information about upcoming events. The thesis includes implementation of native client application for Google Android platform bringing most of the features directly to user's mobile device. The interface between the web application and the mobile application is developed separately and is universally applicable to other existing web application based on Nette Framework. Thanks to this interface, a client mobile application can be created in a matter of minutes and practically without affecting any of the existing web application's source code.
|
|
|
A tool for generating code supporting reading of text files
Černý, Ondřej ; Týnovský, Miroslav (referee) ; Kolovratník, David (advisor)
In the present work we deal with the concept and implementation of tool AutoComp for generating programs for processing structured text files. The tool accepts description of the text whith making use of regular expressions and state machines. The output of the tool is a code in a programming language which gives interface for integrating code to a final program. It is possible to define points for calling specified functions and returning controll to the caller program in the input description. The use of generated code is to tokenize text or parse elementary gramatics. The tool provides two target programming languages and few character encodings. The concept was focused on extensibility to more languages and character encodings.
|
|
|
The Exploitation of Linguistic Information in EBMT
Týnovský, Miroslav ; Žabokrtský, Zdeněk (referee) ; Kuboň, Vladislav (advisor)
Example-based machine translation (EBMT) is a corpus-driven method of machine translation. It builds the translation using analogy of the input text with a translation already made. The benefit of using linguistic knowledge within EBMT is the subject of this thesis. Two language pairs are covered: Czech-English and Czech-German. The thesis covers gathering annotated parallel Czech-German data, design and implementation process of an experimental EBMT system, and the effort to improve it using linguistic knowledge. Detailed evaluation and comparison of both the baseline EBMT and the linguistically enhanced system are described. Evaluation has been done using machine and human evaluation methods. The three automatic evaluation methods are BLEU, NIST and METEOR. The linguistic enhancement of the baseline EBMT system includes comparisons of the input sentence with the examples in the translation memory based on morphology and syntax.
|
guest ::
