The subject of this thesis is non-specific knowledge extraction from the web. In recent years, tools that improve the results of this type of knowledge extraction were created. The aim of this thesis is to become familiar with these tools, test and propose the use of results. In this thesis these tools are described and compared and extraction is carried out using OLLIE. Based on the results of the extractions, two methods of enriching extractions using name entity recognition, are proposed. The first method proposes to modify the weights of extractions and second proposes the enrichment of extractions by named entities. The paper proposed ontology, which allows to capture the structure of enriched extractions. In the last part practical experiment is carried out, in which the proposed methods are demonstrated. Future research in this field would be useful in areas of extraction and categorization of relational phrases.

I study movie studio profit maximization based on an optimization of a political language in the dialogues. I explore the flexibility with which a rational firm slants language of its movies in order to get closer either to a Democratic or a Republican customer. Using computational linguistics I construct vectors of phrase frequency distribution based on a text of almost a decade of U.S. Congress transcripts and 457 randomly chosen movie subtitles. In order to measure distance between the phrase vectors I use chi square statistics and its Monte Carlo approximation. I find no evidence of political slant in movies neither in a movie studio comparison nor for a time-varying comparison of movies in different years. In addition I construct a slant index covering level of political language in a movie. Using the index I find no evidence of impact of political language on movie revenues.

The aim of this master’s thesis is analysis of the multidimensional data. Three dimensionality reduction algorithms are introduced. It is shown how to manipulate with text documents using basic methods of natural language processing. The goal of the practical part of the thesis is to process real-world data from the internet forum. Posted messages are transformed to the numerical representation, then to two-dimensional space and visualized. Later on, topics of the messages are discovered. In the last part, a few selected algorithms are compared.

With the increasing amount of digital data in the form of unstructured text, the importance of natural language processing (NLP) increases. The most suc- cessful technologies of recent years are deep neural networks. This work applies the state-of-the-art methods, namely transfer learning of Bidirectional Encoders Representations from Transformers (BERT), on three Czech NLP tasks: part- of-speech tagging, lemmatization and sentiment analysis. We applied BERT model with a simple classification head on three Czech sentiment datasets: mall, facebook, and csfd, and we achieved state-of-the-art results. We also explored several possible architectures for tagging and lemmatization and obtained new state-of-the-art results in both tagging and lemmatization with fine-tunning ap- proach on data from Prague Dependency Treebank. Specifically, we achieved accuracy 98.57% for tagging, 99.00% for lemmatization, and 98.19% for joint accuracy of both tasks. Best models for all tasks are publicly available. 1

This master's thesis deals with code-switching detection in speech. The state-of-the-art methods of language diarization are described in the first part of the thesis. The proposed method for implementation is based on acoustic approach to language identification using combination of GMM, i-vector and LDA. New Mandarin-English code-switching database was created for these experiments. Using this system, accuracy of 89,3 % is achieved on this database.

The thesis deals with communication of bilingual Czech-German families in North Rhine- Westphalia. We focused mainly on the language of primary school children. The work contains two sections - a theoretical and a practical section. In the first, theoretical part of the work we tried to describe (on the based on literature) the issue of bilingualism, bilingual education and language acquisition. We also discussed the development, more specifically of primary-aged children. Attention in this section is also drawn to the region of North Rhine- Westphalia, where our research was realized. The practical part of the thesis focusses on the qualitative research of communication in multilingual families. We tried to create language biographies of individual families, based on interviews with Czech-speaking parents. Using a questionnaire, we investigated the attitudes of German-speaking parents to bilingual education. We also recorded the linguistic performances of children, which were then analyzed. Our goal was to create the most complete overview of the researched families. Key words language acquisition, bilingualism, younger school age, linguistic biography

This thesis deals with the develompent of Andrej Babiš's Czech. Babiš is a Slovak politician and businessman, who lives in the Czech Republic and whose speeches and media appearances have been part of Czech public discourse for the past few years. The aim of this thesis is to describe how Babiš's language changes over time and how Slovak interference influences it. Hypotheses focus on the number of errors in Babiš's speech, the specific types of errors he makes and the interlingual / intralingual error ratio. The theoretical part of the thesis introduces basic second-language acquisition terminology, which is later used in the practical part. The practical part consists mainly of an error analysis of recordings of interviews with Babiš. Lexical, syntactic, morphological, phonetic, phonological and other differences between Czech and Slovak, which are the source of negative language transfer in Babiš's Czech, are also described. The conclusion shows how parameters of the politician's speech changed over six years and offers an explanation of the change.

The dissertation called the Method of Integration of Valency Syntax into the Czech Language Teaching within Lower Secondary Education deals with the issue of valency syntax in relation to syntax teaching within lower secondary education. The main goal of the dissertation is to describe, specify and define the principles, which serve as the basis for the integration of the valency theory into the Czech language teaching within lower secondary education with regard to pupils' concepts of the structure of a Czech sentence. In its three main parts, the dissertation informs the reader about the development of the perception of the valency syntax as a scientific discipline in the Czech and foreign environment, as the didactic transformation of the curriculum must be always based on scientific theories and comply with them; the second part of the thesis analyses the didactically transformed curriculum of valency syntax in teaching materials intended for lower secondary education. On the basis of a practical quality research, the third part defines pupils' preconcepts related to the formal and semantic structure of a sentence. Among other things, the submitted dissertation points out to the necessity of monitoring the pupils' preconcepts in the Czech language teaching in order to ensure that the teaching is not...

The thesis "Communication and approach to language in multilingual children" is focused on language production of bilingual children, 7 to 13 years old, living permanently in Brussels, the capital of Belgium, who attend the local "Czech school without borders". Its first part presents a theoretical introduction to the field of study, defines all relevant concepts and terms, especially those regarding: child development, the Complexity, Accuracy and Fluency analysis, unprepared speech and other linguistic terms used in following research. In the practical part the empirical execution of aforementioned analysis will be presented. It involves all relevant data about the participants in the research and describes how the interviews were transcribed. The practical part also shows all transcribed interviews of 15 participants of the research. Each participant was monitored for 6 months prior to the interview by the author of the thesis. The abstract of the observation of their language behaviour at school is a part of the practical part as well. This introduction is followed by a detailed Complexity, Accuracy and Fluency analysis and its summary. The end of the thesis provides a table with results of all participants in all parts of the analysis and describes what can be concluded from it. The aim of...

The thesis presents an overview of the layers on which speech develo- pment can be studied and a brief account of the main theoretical branches within these layers. It accounts for the periodization of a child's speech development according to Czech and foreign authors. In the thesis, a syllable is introduced as a phonological unit. Attention is also paid to the related layer of phonotactics; important works concerning the phonotactics of child's speech are mentioned. The target of the experimental part is a longitudinal case study of a child acqui- ring Czech as her mother tongue. For the purpose of the study, a corpus of the child's utterances in her natural environment between the ages of 7 to 27 months was built. The utterances have been analyzed with respect to the frequency of occurrence of different syllable types, with closed syllables and syllables with con- sonant clusters in the center of interest. The found consonant clusters and their reductions have been further analyzed with respect to the findings of the phono- tactic studies mentioned in the theoretical part, in order to confirm or dismiss their relevance for Czech. Most importantly, two theories have been checked, the frames theory commenting on the combinatory potential of the front/middle/back vowels and coronal/labial/dorsal...