|
|
Comparing Prices in Internet Shops
Koukola, Daniel ; Spousta, Miroslav (advisor) ; Semecký, Jiří (referee)
This work deals with problems of comparison shopping engines. It focuses on methods for automatic data extraction from web pages and methods for classification of the extracted data. Several algorithms for solving these problems are proposed and implemented. Also a new algorithm for automatic data extraction based on supervised machine learning techniques is proposed. Implemented methods are evaluated on a test data set and suitable parameters for the tested algorithms are found.
|
|
|
Amarok Speech Control
Kerpl, Lukáš ; Semecký, Jiří (advisor) ; Poch, Tomáš (referee)
In the presented work we study voice control of Amarok music player. The problem of controlling computer using speech is divided into several parts. In the rst part, we need to recognize what user says. In the last one, we should be able to answer to the user or give him some information. In between of these parts, we need to control the program, in our case the Amarok media player. In this work we describe all of these parts, as well as theirs conjunction into working program. Description of the usage of this program and even description of used libraries including potential alternatives is included. The last chapter, Usage and future development, considers the contributions of this program and computer voice control in general at the current time and in the future.
|
|
|
Modular fulltext search for MySQL
Baroš, Martin ; Semecký, Jiří (advisor) ; Podveský, Petr (referee)
An objective of the project is to develop a modular fulltext search engine using MySQL database server. The search engine should operate with the Czech language's specific attributes. There is no endeavor to develop high quality modules solving linguistic problems. Project should provide interface and ability to plug-in (plug-out) next modules. Project's software platform is Unix, programming language C++.
|
|
|
Verb Valency Frames Disambiguation
Semecký, Jiří ; Hajič, Jan (advisor) ; Krbec, Pavel (referee) ; Lopatková, Markéta (referee)
Semantic analysis has become a bottleneck of many natural language applications. Machine translation, automatic question answering, dialog management, and others rely on high quality semantic analysis. Verbs are central elements of clauses with strong influence on the realization of whole sentences. Therefore the semantic analysis of verbs plays a key role in the analysis of natural language. We believe that solid disambiguation of verb senses can boost the performance of many real-life applications. In this thesis, we investigate the potential of statistical disambiguation of verb senses. Each verb occurrence can be described by diverse types of information. We investigate which information is worth considering when determining the sense of verbs. Different types of classification methods are tested with regard to the topic. In particular, we compared the Naive Bayes classifier, decision trees, rule-based method, maximum entropy, and support vector machines. The proposed methods are thoroughly evaluated on two different Czech corpora, VALEVAL and the Prague Dependency Treebank. Significant improvement over the baseline is observed.
|
|
|
Statistical language recognition
Toman, Josef ; Semecký, Jiří (advisor) ; Novák, Václav (referee)
This work deals with the application that uses the machine-learning methods for the automatic language and encoding recognition. Various topics related to the project are analysed here one by one. Brief descriptionof the application should give the reader an idea about the features andthe usage of the program. The n-gram language model, the EM-smoothing algorithm and the entropy is explained in the chapter about the statistical methods. Some implementation matters are subject of analysis, for example the data file format and the inner word representation using the trie. The interpretation of a vast set of experiments evaluating the precision of the application is included as well as a short summary of the possible usage of the software.
|
|
|
Lexical Association Measures Collocation Extraction
Pecina, Pavel ; Hajič, Jan (advisor) ; Semecký, Jiří (referee) ; Baldwin, Timothy (referee)
This thesis is devoted to an empirical study of lexical association measures and their application to collocation extraction. We focus on two-word (bigram) collocations only. We compiled a comprehensive inventory of 82 lexical association measures and present their empirical evaluation on four reference data sets: dependency bigrams from the manually annotated Prague Dependency Treebank, surface bigrams from the same source, instances of surface bigrams from the Czech National Corpus provided with automatically assigned lemmas and part-of-speech tags, and distance verb-noun bigrams from the automatically part-of-speech tagged Swedish Parole corpus. Collocation candidates in the reference data sets were manually annotated and labeled as collocations and non-collocations. The evaluation scheme is based on measuring the quality of ranking collocation candidates according to their chance to form collocations. The methods are compared by precision-recall curves and mean average precision scores adopted from the field of information retrieval. Tests of statistical significance were also performed. Further, we study the possibility of combining lexical association measures and present empirical results of several combination methods that significantly improved the performance in this task. We also propose a model...
|
|
|
Unsupervized construction of semantic lexicon
Kirschner, Martin ; Semecký, Jiří (advisor) ; Spousta, Miroslav (referee)
In present work are studied possibilities of unsupervized constuction of semantic lexicons. There are compared advantages and disadvantages of several annotation methods. More closely developed is one solution, which is based on latent semantic analysis (LSA). Unlike the others, this method solves the problems of polysemy and synonymy. Below are described details of the algorithm implementanion and evaluation. This thesis provides a set of tools for construction, browsing and conversion of semantic lexicons for czech language. Used methods do not depend on the language.
|
|
|
Sport Manager Simulator
Zákutný, Roman ; Novák, Václav (advisor) ; Semecký, Jiří (referee)
The first part of presented work contains a description of the technological structure of the web application - a system for virtual simulation of a sport world controlled by user's active management of a team. The structure is divided into multiple layers - presentation, appliction and database. Use and utilization of particular frameworks on these layers are described in the work. In the second part we examine the implementation of the model used to generate the course of the match taking into account the parameters influenced by the managing abilities of users owning their competing teams. We introduce Markov models, their use in our model, connection between coincidence and observation of the situations and we perform statistical analysis of the results.
|
|
|
Ohníček;changes in structure of magazine and in text characteristics between 1950-2001
Semecký, Jiří ; Končelík, Jakub (advisor) ; Wolák, Radim (referee)
The undergraduate thesis entitled "Časopis Ohníček: kvantitativní obsahová analýza vybraných pozornostních prvků 1950-2001" (Ohníček Journal: Content analysis of representation of selected attention-catching features in 1950-2001) deals with representation and role of visual attention-catching features in a children's printed periodical. The thesis describes a brief history and context of publishing the journal and with sensory perception of human organism with stress on visual perception, in connection with the process of attracting inadvertent as well as intentional attention. By means of quantitative content analysis it attempts at capturing the development of extent of representation of selected features with graphic significance in the journal, in particular of headlines and pictorial components, which immediately affect visual perception of the recipient and which are thus endowed with the largest potential to capture inadvertent attention through orienting responses. Such inadvertent attention may subsequently transform into intentional attention on the basis of the level of meeting individual expectations and interests. External form of respective contributions thus assists in awakening interest in their content. The development of the selected attention- capturing features is monitored...
|
|
|
Verb Valency Frames Disambiquation
Semecký, Jiří
Semantic analysis has become a bottleneck of many natural language applications. Machine translation, automatic question answering, dialog management, and others rely on high quality semantic analysis. Verbs are central elements of clauses with strong influence on the realization of whole sentences. Therefore the semantic analysis of verbs plays a key role in the analysis of natural language. We believe that solid disambiguation of verb senses can boost the performance of many real-life applications. In this thesis, we investigate the potential of statistical disambiguation of verb senses. Each verb occurrence can be described by diverse types of information. We investigate which information is worth considering when determining the sense of verbs. Different types of classification methods are tested with regard to the topic. In particular, we compared the Na¨ive Bayes classifier, decision trees, rule-based method, maximum entropy, and support vector machines. The proposed methods are thoroughly evaluated on two different Czech corpora, VALEVAL and the Prague Dependency Treebank. Significant improvement over the baseline is observed.
|
guest ::
RSS feed.