|
|
Extrakce znalostních grafů z projektové dokumentace
Helešic, Tomáš ; Nečaský, Martin (advisor) ; Kruliš, Martin (referee)
Title: Knowledge Graph Extraction from Project Documentation Author: Bc. Tomáš Helešic Department: Department of Software Engineering Supervisor: Mgr. Martin Nečaský, Ph.D. Abstract: With the new research progress in the natural language processing and information extraction from text, new possibility of automatic knowledge acqui- sition and its grouping into Knowledge graphs, that are catching the semantic relations between these entities is emerging. For these Knowledge graphs, data storages and also query languages already exists, which allow more precise and relevant search possibilities compare with current full text search engines. The goal of this thesis is to explore the opportunity of automatic extraction of infor- mation from project documentation with the use of linguistic text processing, design a proper data storage and build a search engine over it. Keywords: Knowledge grahs, Information extraction, Natural language process- ing, Resource Description Framework 1
|
|
|
How to Create Self-Driven Education: The Social Web & Social Sciences, Coursera & Khan Academy 2014 Case Study
Růžička, Jakub ; Remr, Jiří (advisor) ; Soukup, Petr (referee)
This diploma thesis is concerned with the possibilities of the social web data employment in social sciences. Its theoretical part describes the changes in education in the context of the dynamics of contemporary society within three fundamental (interrelated) dimensions of technology (the cause and/or the tool for the change), work (new models of collaboration), and economics (sustainability of free & open-source business models). The main methodological part of the thesis is focused on the issues of sampling, sample representativeness, validity & reliability assessment, ethics, and data collection of the emerging social web research in social sciences. The research part includes illustrative social web analyses and conclusions of the author's 2014 Coursera & Khan Academy on the Social Web research and provides the full research report in its attachement to compare its results to the theoretical part in order to provide a "naive" (as derived from the social web mentions and networks) answer to the fundamental question: "How to Create Self-Driven Education?" Powered by TCPDF (www.tcpdf.org)
|
|
|
Metrics for Optimizing Statistical Machine Translation
Macháček, Matouš ; Bojar, Ondřej (advisor) ; Popel, Martin (referee)
State-of-the-art MT systems use so called log-linear model, which combines several components to predict the probability of the translation of a given sentence. Each component has its weight in the log-linear model. These weights are generally trained to optimize BLEU, but there are many alternative automatic metrics and some of them correlate better with human judgments than BLEU. We explore various metrics (PER, WER, CDER, TER, BLEU and SemPOS) in terms of correlation with human judgments. Metric SemPOS is examined in more detail and we propose some approximations and variants. We use the examined metrics to train Czech to English MT system using MERT method and explore how optimizing toward various automatic evaluation metrics affects the resulting model.
|
|
|
Authorship Identification
Fabiánek, Ondřej ; Škoda, Petr (referee) ; Smrž, Pavel (advisor)
This bachelor's thesis deals with authorship identification based on knowledge of author's previous texts. The aim is to analyze existing methods of authorship attribution and create a system, which is capable of highly successful authorship identification. The system is based on a multivariate analysis and specializes at English books. Part of the solution is also a graphic user interface.
|
|
| |
|
|
Entity Knowledge Base Creation from Czech Wikipedia
Sychra, Martin ; Otrusina, Lubomír (referee) ; Smrž, Pavel (advisor)
The aim of this thesis is to propose and implement a system for an automatic extraction of named entities from Czech Wikipedia, to create a knowledge base consisting of these entities and to evaluate results of the created system. The first part explains basic notions of this field and discusses related work. The main part proposes several methods of extraction and details their implementation. The following types of entities are extracted: people, places, events and organizations. The final part of the thesis presents results, i.e., the success of the individual methods for each entity type and statistics on extraction of the individual entities in the whole Czech Wikipedia context.
|
|
|
ChatBot Based on Language Modelling
Radvanský, Matěj ; Szőke, Igor (referee) ; Skála, František (advisor)
This paper addresses the use of language modeling using neural networks in the chatbot. Problem is solved by using natural language processing and first step of generating response based on user input is input analysiss. As next beginings of sentences are created which are completed by output of neural network. All created sentences form final chatbot response. There was a comparisson with chatbot Cleverbot and measure of intelligence for both chatbots was determined. Based on testing results, some techniques for future progress were concluded.
|
|
|
Semantic Similarity of Articles
Veselovský, Martin ; Otrusina, Lubomír (referee) ; Kouřil, Jan (advisor)
This bachelor's thesis deals with modelling of structure of semantic relationships among articles in English language. There are introduced existing methods of articles representation and computation of similarity. The base method is vector space model, which represents document as vector of words. There are given weights of importance to these words using TF-IDF method. Next, there are described advanced methods of modelling, Latent semantic analysis (LSA) and Latent Dirichlet allocation (LDA). This thesis also deals with articles, which are semantically annotated, while weights of annotation words are computed by Stochastic Gradient Descent method. Evaluation of results takes place on the prepared test corpus of documents to which there is reference similarity evaluation.
|
|
|
Dictionary Up-Translation System
Schovajsa, Michal ; Kouřil, Jan (referee) ; Smrž, Pavel (advisor)
The thesis concerns with the processing of dictionaries in electronic form, converting them into an unified form, and the problems arising in the process in particular. The subject of the work is to create a system for the elimination of some of these problems in order to facilitate machine processing of dictionaries. At first, different issues of dictionaries transferred into an unified form are concerned. Then, the thesis deals with the solution of these issues and the creation of tools for this purpose. Finally, the results and the efficiency of the instruments created are evaluated.
|
|
|
Machine-Learning in Natural Language Processing
Otrusina, Lubomír ; Šilhavá, Jana (referee) ; Smrž, Pavel (advisor)
This beachelor's thesis deals with word sense disambiguation problem using the machine learning techniques. There are shortly presented problems of word sense disambiguation and its timeline. There are described methods and approaches, especially the naive Bayes classifier that is implemented in the system. There's illustrated a simple example of using this classifier. In a practical section is described project of system based on naive Bayes classifier including description of various algorithms used in the system. Finally there are described evaluation and analysis of the system. This created system took part in an international competition on semantic evaluation workshop SemEval-2007.
|
guest ::
