|
| |
|
|
Automatic combinations of feature templates
Dubovský, Jakub ; Vidová Hladká, Barbora (referee) ; Novák, Václav (advisor)
Searching for useful combinations of features and feature templates is not a simple task. Though combination is valuable tool for increasing accuracy of machine learning. This paper tries to suggest an algorithm for automatic search for useful combinations of categorical features and their templates. An attempt to use simulated annealing and modified genetic algorithm for search process is studied. Construction of evaluation function for assessing categorical feature template is present as well. Features and feature templates are combined separately and together. The best increase of accuracy reached by suggested procedures on datasets used is around 0.1 percentage points. Experiments were made just on two datasets. Thus further testing of algorithm on other datasets is needed to verify its usefulness in general. However experiments indicate that it can be considered as a base of usable algorithm. Simple command-line application is part of work. It was developed and used for experimentation.
|
|
|
Statistical language recognition
Toman, Josef ; Novák, Václav (referee) ; Semecký, Jiří (advisor)
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.
|
|
|
Automatický word alignment
Kravalová, Jana ; Novák, Václav (referee) ; Pecina, Pavel (advisor)
Word alignment is a crucial component of modern machine translation systems. Given a sentence in two languages, the task is to determine which words from one language are the most likely translations of words from the other language. As an alternative to classical generative approach (IBM models) new methods based on discriminative training and maximum-weight bipartite matching algorithms for complete bipartite graphs have been proposed in recent years. The graph vertices represent words in the source and target language. The edges are weighted by measures of association estimated from parallel training data. This work focuses on the effective implementation of maximum weight bipartite matching algorithm, implementation of scoring procedures for graph vertexes, and basic experiments and their evaluation.
|
|
|
Automatical diacritics reconstruction for mobile devices
Železník, Jaroslav ; Novák, Václav (advisor) ; Češka, Pavel (referee)
The goal of this thesis is to develop a Java MIDP application for automatic reconstruction of the diacritics in Czech and Slovak texts. The application is intended for mobile devices and the size of the statistical model must meet the limitations imposed by the devices. A separate Java application will create the model files used by the mobile application. The size of the model can be adjusted to fit a specific device.
|
|
|
Clients Management Database for Tax Advisors
Štěrba, Jan ; Novák, Václav (referee) ; Bojar, Ondřej (advisor)
The goal of this project is to implement a system for management of tax advisor's clients. The purpose of this system is to manage all relevant data about clients and work assignments and to simplify time management. An important feature of the system is to generate summaries of assignments, schedule and deadlines and also give statistical data about payments to the advisor. The system will provide output to external data formats and simultaneous work of multiple users.
|
|
|
Logical Representation of Language Structures
Novák, Václav
This thesis is meant as a basis for a semantic annotation manual. The annotation will translate the logical structure of written and spoken speech by a language-independent manner. The basis for this thesis can be found in temporal logic of Mark Steedman and the formalism introduced here elaborates on Patrick Blackburn's hybrid multimodal logic and Dependency Grammar Logic of Geert-Jan Kruijff, although it represents not the level of linguistic meaning but rather the level of logical meaning (content). The main issues discussed in this thesis are input text segmentation into logical units, discourse referents representation, time-spatial representation, the impact of topic-focus articulation on the logical content and the semantic categories needed for the annotation. In the last two chapters a complete discourse example and the annotation of 37 test cases of Jan Hajič are presented.
|
|
|
Influence of application of activators of vital soil functions on selected soil characteristics at ZS Sloveč, a.s.
Novák, Václav ; Šařec, Petr (advisor) ; Šařec, Ondřej (referee)
This thesis covers issues of soil physical properties changes after the application of farmyard manure, industrial fertilizers or substances which aim for improvement of vital soil properties. The first part of the thesis is focused on soil issues and besides other things it covers types of soil degradation or physical soil properties. In the next part of the thesis there is a focus on substances which improve physical soil properties. Biochar and other representative products of company PRP Technologies are described in this part. The practical part of this thesis describes selected properties changes which have arisen after a year of field trials in company Zemědělská společnost Sloveč, for example changes in traction or penetration resistance. The last part includes the summary of results, conclusion and the recommendation of author.
|
|
|
Living in Old Brno - architecture and urban study
Novák, Václav ; Sátora, Josef (referee) ; Kopáčik, Gabriel (advisor)
The aim of this thesis is to find new forms of housing for socially excluded inhabitants of the Czech Republic. The proposed set of two buildings is located in the street Kopečná in the district of Staré Brno. The buildings are placed in the immediate vicinity and their functions are independent of each other. The first building provides both acute and long-term assistance and temporary accommodation for homeless people. Inside this building is a low threshold day centre, night shelter and sheltered housing. There is a hierarchy of different levels of sheltered housing created by specific dispositions, which cover the wide diversity of needs of the target group. The second object is a residential building providing accommodation in the concept of "housing first" ("housing first"). Inside the building there are sheltered workshops (cafe, bakery and bicycle repair shop) provides the employability skills and social flats serves as a pre independent living.
|
|
|
Detect the presence of obesity of children 10-11 years old in České Budějovice by BMI method
NOVÁK, Václav
The main aim of this diploma thesis is to determine the occurrence of obesity in schoolchildren attending primary schools in České Budějovice. The research focused on determining whether students of the 5 th grades aged 10 and 11 are obese based on their body mass index, age and sex BMI percentile graph. the findings were compared against the Parizkova and Lisa method (2007). It was discovered that 5,13 % of the responded girls and 6,67 % of the boys are obese. The results suggest that out of 195 participants, 11,5% of them suffer from obesity and that the occurrence of obesity is more frequent in boys.
|
guest ::
RSS feed.