|
Linguistic Text Compression
Kazík, Ondřej ; Dědek, Jan (referee) ; Lánský, Jan (advisor)
The compression of texts written in natural language can exploit information about its linguistic character. It is shown that separation of coding of part-of-speech tags of a sentence (type of sentence) from the text alone can improve resulting compression ratio. For this purpose the tagging method NNTagger based on neural networks is designed. This thesis is focused on speci fication of the compression model of texts written in Czech. We propose methods of constructing of initial dictionaries and test their influence on the compression ratio.
|
|
The first czech Teacher´s Institute in South-Bohemian. Teacher´s Institute in Soběslav in years 1871 - 1948
TRNÍKOVÁ, Věra
Presented diploma thesis is based on the study of scientific literature concerning the topic of the work and on the source documents preserved in the State Regional Archives in Tábor. The objective of the work is to define the establishment and the functioning of the Teacher´s Institute in Soběslav, which was the oldest South-Bohemian school of its type providing education to the public interested in teaching at elementary schools in the Czech language. Pegagogium founded in 1871 in Tábor and moved into the picturesque town of Soběslav a year later provided, during seventy-seven years of its functioning, a quality education and the possibility to take the final school-leaving exam to more than four thousand of future teachers of elementary and council schools. Following pages deal with the characteristics of these people and also with the destinies of the most significant teachers, their educators and the school which provided them with the necessary knowledge. The work also presents information concerning the everyday life of the school, extracurricular activities and the relation of the school towards the host town during the changes in the Czech countries at the end of the 19th and the beginning of the 20th centuries. The work aims not only to describe the history of the Teacher´s Institute but also to enrich the knowledge of the position of the South Bohemian schools in the whole-republic educational network and to specify the changes of the teacher´s education from the period of Maria Theresa?s reforms to the first half of the 20th century.
|
|
Knowledge management and its application in human resource management
Běhounková, Olga ; Tureckiová, Michaela (advisor) ; Beneš, Milan (referee) ; Plamínek, Jiří (referee)
This doctoral thesis proposes to specify a relatively new managerial discipline - Knowledge Management (KM) in a different way than it is done in an accessible literature. Nowadays, KM represents a wide and intensively discussed topic with many critics as well as advocates. Above all, it is a practical discipline that came out of practice and it is inseparably connected with it. Intangible (knowledge) assets are its main essence. KM is a complex process that must be supported by a strong foundation of enablers. The enablers for KM are corporate strategy and culture, team work, informal networking, measurement, and technology. Each of these must be designed and managed together with the others and supported by organizational process. The process usually involves several of the following stages or sub-processes in the use of knowledge: to create, identify, formalize, organize, disseminate, share, adapt, use and evaluate it. Effective implementation of KM brings the right knowledge to the right associates on time and in the right form to the organization. The thesis consists of four chapters. The first and second part are focused on the theoretical definitions of KM bases and terminology, explicitly "organization", "management", "manager" and "knowledge", i. e. from different point of view together with an...
|
|
Community portal
Rosík, Jiří ; Nekvasil, Marek (advisor) ; Schenk, Jan (referee)
The aim of this work is to map the sphere of community portals, to determine what are the characteristics of users and how they behave. Subsequently, this knowledge confront with the current state. In the sphere of semantic web technologies are still solved the problem of their introduction into practice. The work aims to describe these technologies and find out how it can be in the field of social networks applied and where can they find the application. For those, who want to start their own small community portal is designed solution in the form of simple application, based on free available technologies. Here is the aim to provide warp and inspiration to set up a similar project. Taken into account is the economics, which is intended to outline the sphere of financing community portal in its infancy. At the end we specify problems and difficulties faced by entrepreneurs of existing sites. The main contribution consists of an analysis of studies on users of community portals and deduction the knowledge about behavior and characteristics of these participants. You can apply this information for marketing or any other work in the field of social networks. The contribution of the practical part of the work lies in exploring the applicability of semantic technologies in social networks and the subsequent implementation of any proposal. For those interested in establishing the portal is a useful proposal for their own applications, including economic aspects.
|
| |
|
Single-Image License Plate Recognition
Vala, Tomáš ; Winkler, Zbyněk (referee) ; Štanclová, Jana (advisor)
The diploma thesis deals with analysis and design of an automatic system for a single-image license plate (LP) recognition. Input data came from a system for car speed measurement on the roads in the emirate of Abu Dhabi (the United Arab Emirates). A goal is to recognize a numeric identi cation code on a LP and a type of the LP. In details there are discussed individual parts of the system for LP recognition. A particular stress is put on segmentation and grouping of individual characters to be able to utilize specific information on the LP from Abu Dhabi during the character recognition. The diploma thesis compares several methods for character recognition - a minimum distance classi er, feed-forward neural networks and hidden Markov models. The methods (and some modifications of them) are tested on real data and compared according to the percentage of correctly recognized digits (at fi rst the digits were manually classi ffied). Wrongly recognized LPs are undesirable. Instead in some cases it is better to mark the LP as "unrecognized". That's why there were introduced sophisticated tests for recognized digits reliability veri cation. A summary of our own results is included.
|
|
Customer realationship management and nets of independent applications
Koňas, Jakub ; Král, Jaroslav (advisor) ; Yaghob, Jakub (referee)
The objective of this thesis is to study the concept of software confederations (networks of independent applications, NIA) orientated on its utilization in the field of customer relationship management (CRM) and the enterprise information systems. The opening chapters of the thesis specify the meaning of CRM, present the evolvement of CRM systems and clarify their internal structure and provided features. The insight into the CRM subject terminates with extensive survey of CRM systems which are being sold within the Czech Republic. The following chapters explain the conception of NIA and evaluate advantages and disadvantages of contemporary technologies for building NIA. The equivalence between NIA and SOA-based solutions is recognized. The final chapters discuss the speciffic limitations and possible benefitss of NIA in the CRM fields.
|
|
Fog computing
Poplštein, Karel ; Karkošková, Soňa (advisor) ; Feuerlicht, Jiří (referee)
The purpose of this bachelor's thesis is to address fog computing technology, that emerged as a possible solution for the internet of things requirements and aims to lower latency and network bandwidth by moving a substantial part of computing operation to the network edge. The thesis identifies advantages as well as potential threats and analyses the possible solutions to these problems, proceeding to comparison of cloud and fog computing and specifying areas of use for both of them. Finally, it examines the perspective of fog computing technology and its possible future development.
|
|
Modernization of compressor station
Kocourková, Markéta ; Dlouhý, Martin (advisor) ; Dlouhá, Zuzana (referee)
The thesis is concerned with multi-criteria evaluation of alternatives of modernization of compressor station for natural gas. This thesis is divided into practical part and theoretical part. The theoretical part will introduce historical development of the gas industry and the current state of the gas network in our country. The following describes a method for operating a compressor station, its technology units and existing operational processes. The theoretical part is a chapter that describes methods of multi-criteria evaluation of alternatives and a way of specifying the methods suitable for constructing weights of the criteria. The fifth chapter contains an application example, the aim of which is to choose the suitable way of modernization of the compressor station that would meet the requirements of planned projects. Creating recommendations for the management of the company fulfilled the goal of this thesis.
|
|
The Utilization of Soft Computing in Ordering Cycle Management
Šustrová, Tereza ; Fecenko, Josef (referee) ; Jurová, Marie (referee) ; Marček, Dušan (referee) ; Dostál, Petr (advisor)
This doctoral thesis deals with possibilities of using advanced methods of decision-making - Soft Computing, in company’s ordering cycle management. The main aim of the thesis is to propose an artificial neural network model with an optimal architecture for ordering cycle management within the supply chain management. The proposed model will be employed in an organization involved in retailing to ensure smooth material flow. A design and verification of artificial neural networks model for sales prediction is also part of this doctoral thesis as well as a comparison of results and usability with standard and commonly used statistical methods. Furthermore, the thesis deals with finding a suitable artificial neural network model with architecture capable of solving the lot-size problem according to specified inputs. Methods of statistical data processing, economical modelling and advanced decision-making (Soft Computing) were utilized during the model designing process.
|
guest ::
