|
|
Analysis and visualization of OCR output
Nová, Kateřina ; Vidová Hladká, Barbora (advisor) ; Mírovský, Jiří (referee)
Optical Character Recognition (OCR) is a process of converting text from images to a machine-readable text. We run three OCR systems (Tesseract, Ocrad and GOCR) on an original multilingual OCR dataset and perform statistical and linguistic analysis of the results in order to compare the tested systems and investigate typical OCR errors. 1
|
|
|
Syntactically-based classification of Czech sentences
Kríž, Vincent ; Vidová Hladká, Barbora (advisor) ; Mírovský, Jiří (referee)
Classification of syntactically meaningful sentences is a very useful task for the applications of natural language processing, for example machine translation, search engines and question answering systems. The theoretical linguistic research considers the language to be a system of layers. In our project, a term 'to-be-meaningful' will be specified with respect to this point of view. Namely, the morphological and syntactic layers will be considered. A knowledge-based algorithm classifying a string of Czech words being either meaningful or meaningless will be proposed and implemented. Before being classified, strings will be pre-processed by the external modules. Czech will be used as the object language.
|
|
|
Automatic Recognition of Mouse Gestures for Application Controlling
Bedecs, Vladimír ; Homola, Petr (advisor) ; Mírovský, Jiří (referee)
In the present work we study problems of automatic recognition of mouse gestures for application controlling. This includes creation of program, which is able to recognize already defined gestures made by mouse and with them provides controlling of applications. The program is written or operating system Windows. This work includes description of some algorithms for recognition of mouse gestures, which are used in enclosed program Mouse Gestures. Also analyzes hook techniques of operating system and uses them for application controlling. An important chapter of this work is processing of gesture's meaning, which includes gesture's meaning structures implementation, so that the processing of gesture's meaning can be most effective. In the last pages of this work can reader find some advices and tutorials for using made application.
|
|
|
An Implementation of the Extended Game Quoridor
Trojánek, Tomáš ; Mírovský, Jiří (advisor) ; Zeman, Daniel (referee)
This thesis analyses a classic board game called Quoridor and subsequently its implementation into a form of computer multiplayer application. At first, rules and its extensions from offical version are described, namely - new type of obstacles and generalization of the game board so that it allows bigger amount of players. These specifics alter the game and bring new challenges. A description of algorithms and mathematical study that solves these problems effectively folow. The described version is compared to the original. Finally thesis write about the design, realization and important mechanisms of the application, which contains, among others, implementations of individual algorithms from this text. Program allows to test proved theorems in real environment. It is a network multiplayer game with possibility to add computer oponents.
|
|
|
Machine Learning Approach to Anaphora Resolution
Novák, Michal ; Žabokrtský, Zdeněk (advisor) ; Mírovský, Jiří (referee)
Anaphora resolution is the key task for some of the Natural Language Processing (NLP) tasks like the information extraction or dialog systems. It can be also valuable in machine translation. All the previous works concerning the anaphora resolution in Czech language mostly focused on the pronoun coreference. Thanks to the recent project of the annotation of extended anaphoric relations in Prague Dependency Treebank 2.0 his work goes further. It attempts to resolve noun phrase coreference, identity-of-sense anaphora and part-whole bridging relations. It has adopted some of the stateof-the-art approaches in the area of machine learning approaches to anaphora resolution, particularly the ranking and the joint anaphor identi cation with the antecedent selection. It introduced a plenty of features and analyzed their contribution on the success rate. The best model of noun phrase coreference achieves the F-score of 39.4%.
|
|
|
Netgraph-A Tool for Searching in the Prague Dependency Treebank 2.0
Mírovský, Jiří ; Hajič, Jan (advisor) ; Rosen, Alexandr (referee) ; Ondruška, Roman (referee)
Three sides existed whose connection is solved in this thesis. First, it was the Prague Dependency Treebank 2.0, one of the most advanced treebanks in the linguistic world. Second, there existed a very limited but extremely intuitive search tool - Netgraph 1.0. Third, there were users longing for such a simple and intuitive tool that would be powerful enough to search in the Prague Dependency Treebank. In the thesis, we study the annotation of the Prague Dependency Treebank 2.0, especially on the tectogrammatical layer, which is by far the most complex layer of the treebank, and assemble a list of requirements on a query language that would allow searching for and studying all linguistic phenomena annotated in the treebank. We propose an extension to the query language of the existing search tool Netgraph 1.0 and show that the extended query language satisfies the list of requirements. We also show how all principal linguistic phenomena annotated in the treebank can be searched for with the query language. The proposed query language has also been implemented - we present the search tool as well and talk about the data format for the tool. An attached CD-ROM contains the installation of the tool.
|
|
|
Czech morphological guesser
Suchánek, Michal ; Hlaváčová, Jaroslava (advisor) ; Mírovský, Jiří (referee)
The first step of text analysis is tagging word forms with morphological tags. These tags describe the part of speech, person (where applicable), number, etc. This information is used for further analysis of the text. Tags are automatically assigned by looking up the word form in the morphological dictionary. This gives good results for the Czech language because the word forms express the morphological categories to some extent. Unlike English words, Czech words often change their ending when their morphological category changes. Words that are not present in the dictionary can be tagged with a guesser. The guesser described here uses the similarity of unrecognized words with words already present in the dictionary.
|
|
|
System of Czech numerals and their automatic recognition in texts
Bureš, Jan ; Hlaváčová, Jaroslava (advisor) ; Mírovský, Jiří (referee)
This thesis has two main goals. The first goal is systematic classification of Czech numerals and other quantitative phrases (including multiple-word) with special regard for possible use during automatic recognition of Czech text. The main source of data for theis classification is current Czech grammar and author's research in Czech language corpora The second goal is development of tool for automatic recognition of numerals in Czech text based on the system developed during the first phase of this thesis. This includes determining basic morphological attributes of numerals and their numeric value, where possible and applicable. The tool is even prepared to deal with the fact, that the gramatic rules for numerals are often disregarded.
|
|
|
Automatic Resolution of Pronoun Coreference in Czech
Košarko, Ondřej ; Mírovský, Jiří (advisor) ; Vidová Hladká, Barbora (referee)
Title: Automatic Resolution of Pronoun Coreference in Czech Author: Ondřej Košarko Department: ÚFAL MFF UK Supervisor: RNDr. Jiří Mírovský, Ph.D. Supervisor's email address: mirovsky@ufal.mff.cuni.cz Abstract: The aim of this thesis is to introduce a procedure for automatic pronomial coreference resolution in Czech texts. The text is morphologically and analytically annotated acording to the system of Prague Dependency Treebank. The procedure uses a machine learning method; for its training a set of manually annotated data from Prague Dependency Treebank is used. Evaluation of the results is also part of this thesis. Keywords: pronomial coreference, automatic resolution, machine learning
|
|
|
Czech prefixes
Hrušecký, Michal ; Hlaváčová, Jaroslava (advisor) ; Mírovský, Jiří (referee)
Automatic recognition of new prefixes in Czech language is studied in the presented work. Several methods of automatic recognition of prefixes are described and one of them is analyzed more deeply. Analyzed method is also implemented in the software which is part of this work. Software can be found on attached CD including source code and example datasets. CD includes also results of all tests mentioned in the presented work.
|
guest ::
RSS feed.