|
Aplikace dataminingových metod na bankovní data
Melichar, Miloš
The thesis deals with pre-processing of two data sets with information on clients, loans and debit cards. The data sets were separately pre-processed and modeled by SPSS Modeler using a number of methods and algorithms. For the modeling purposes, three classification data mining tasks were defined: loan approving or rejecting, loan rating and debit card type assignment. By using the selected methods of machine learning techniques the classification models were built for each task. Models accuracy was tested by script written in SPSS language for automation. All tasks were supplemented by clustering technique based on latent factors gained by factor analysis. Factor analysis combined with clustering presents another approach in pattern discovery.
|
|
Smartcard authentication
Juras, Stanislav ; Burda, Karel (referee) ; Hajný, Jan (advisor)
The master’s thesis outlines the problem of authentication. It describes authentication factors like ownership, knowledge and inherent. There are described properties of each of that. Authentication based on ownership focuses on authenticators - the smartcards. The thesis also describes different types of smartcards (contact, contactless and hybrid smartcards) and refers to their basic properties. Emphasis is placed on the description of contact and contactless smartcard, specifically focusing on .NET smartcards. It describes their internal components such as memory (RAM, ROM, EEPROM), crypto processor etc. Smartcards are also examined in terms of support for cryptographic primitives. The thesis also introduces the cryptographic methods and principles describing symmetric and asymmetric ciphers. Symmetric ciphers are divided in to stream and block ciphers. There is description of asymmetric cipher, digital signature etc. This work also touches on the fundamental principles required for safe programming. Part of this work is also practical implementation (programs). Practical part aims to implement the communication between the user and AC (Access Control) AASR system. The first suite of applications provides encrypted communication between the PC and smartcards. These applications should create on smartcard services that will be able to perform operations on the client side, which are necessary to authenticate in the AASR system. The paper also presents algorithms for working with big numbers - addition, subtraction, multiplication, and Montgomery's algorithm for multiplication. The second application implements the functionality of AC components (Access Control). This functionality is for example – authenticate received token, digital signature authentication, generating random numbers, logging etc.
|
|
Potential calculation of mutual information from a time series
Hubr, Ivo ; Smékal, Zdeněk (referee) ; Mekyska, Jiří (advisor)
Mutual information is one of the factors used in traffic analysis and preparation phase space. Begin of this work deal with information theory, focusing on the calculation of mutual information. To calculate this parameter has been available for many algorithms which are analyzing in this final work. Two of the algorithms (Fraser-Swinney and calculation of mutual information using adaptive XY subdivision) are applied to the input data Rössler’ attractor, as shown in the output tables and graphs. The third consideration method is the computational Dinh-Tuan-Pham algorithm. The main goal of this work is a comparison of efficiency, speed and accuracy of the calculation of these algorithms.
|
|
Software support of education in cryptography based on integer factorization
Vychodil, Petr ; Martinásek, Zdeněk (referee) ; Burda, Karel (advisor)
This thesis deals with new teaching software, which supports asymmetric encryption algorithms based on the issue of large numbers´ factorization. A model program was created. It allows to carry out operations related to encryption and decryption with an interactive control. There is a simple way to understand the principle of the RSA encryption method with its help. The encryption of algorithms is generally analysed in chapters 1,2. Chapters 3 and 4 deals with RSA encryption algorithm in much more details, and it also describes the principles of the acquisition, management and usage of encryption keys. Chapters 5 suggest choosing of appropriate technologies for the creation of the final software product, which allow an appropriate way to present the characteristics of the extended RSA encryption algorithm. The final software product is the java applet. Aplet is described in the chaprers 6 and 7. It is divided into a theoretical and practical part. The theoretical part presents general information about the RSA encryption algorithm. The practical part allows the users of the program to have a try at tasks connected with the RSA algorthm in their own computers. The last part of Java applet deals with the users´ work results. The information obtained by the user in the various sections of the program are satisfactory enough to understand the principle of this algorithm´s operations.
|
|
Investigative CT algorithms for the early diagnosis of ischemic brain lesions
VOHNOUTOVÁ, Markéta
Stroke is the second most often cause of death in the Czech Republic after heart attack. It is not to say that older people are afflicted in particular. In the last period of ten years there are more often cases of stroke and ictus for people at the age of 40. The aim of this bachelor work is processing of knowledge acquisition in summary. The main point of this work is to evaluate the asset of new software programmes of multidetector computed tomography during early diagnostic of sudden strokes. In part of Present state there are following terms described in detail. It is etiology, aethiopathogenesis, risk factors of stroke. The next part is centred on anatomy of brain vascular supply and computed tomography Nowadays, software programmes have been improved enough for early diagnostics of ischaemic brain lesion. It is caused with continual improvement of diagnostic imaging and its digitalization. In many medical fields it has been spoken about new field of imaging - digital radiography. The computed tomography represents star part in the diagnostics, scanning of non-enhanced brain is the first period of the diagnostics, this algorithm shows us if it concerns haemorrhage or non-haemorrhage centre, continuing perfusion of brain which represents chat percent of tissue is damaged forever and it is finished with diagnostic process Angiography. The following part is focused on processing of data acquisition from České Budějovice Hospital, Inc. and Masaryk Municipal Hospital in Jilemnice. In these hospitals there are the numbers of patients examined CT compared. My hypotesis is confirmed and I can say that usage of new software programmes for early diagnostic of sudden strokes reduces per cent of complications during their cure.
|
|
Improvement of CO2 eddy fluxes modelling in topographically complex terrain
Šigut, Ladislav
Annual sums of net ecosystem exchange (NEE) estimated by eddy covariance method (EC) are often used for comparisons among sites. But application of EC in topographically complex terrain restrains application of standard methods for their estimation. Analysis of friction air velocity response of night-time CO2 fluxes and estimation of mean daily ecosystem respiration from daytime NEE showed that night-time data and fluxes measured early after sunrise or late before sunset are not sufficiently reliable for computation of accurate annual sums of NEE. Therefore presented method takes into account auxiliary soil chamber measurements and applies correction factor in ecosystem respiration model calibrated to biomass inventory results. Also automated algorithm for computation of light response curve parameters in constrained range was established. This method produced accurate annual sums of NEE that were only 7.2% ± 5.2 higher than values determined by biomass inventory method.
|
|
Multi-channel Methods of Speech Enhancement
Zitka, Adam ; Balík, Miroslav (referee) ; Smékal, Zdeněk (advisor)
This thesis deals with multi-channel methods of speech enhancement. Multichannel methods of speech enhancement use a few microphones for recording signals. From mixtures of signals, for example, individual speakers can be separated, noise should be reduced etc. with using neural networks. The task of separating speakers is known as a cocktail-party effect. The main method of solving this problem is called independent component analysis. At first there are described its theoretical foundation and presented conditions and requirements for its application. Methods of ICA try to separate the mixtures with help of searching the minimal gaussian properties of signals. For the analysis of independent components are used different mathematical properties of signals such as kurtosis and entropy. Signals, which were mixed artificially on a computer, can be relatively well separated using, for example, FastICA algorithm or ICA gradient ascent. However, difficult is situation, if we want to separate the signals created in the real recording enviroment, because the separation of speech people speaking at the same time in the real environment affects other various factors such as acoustic properties of the room, noise, delays, reflections from the walls, the position or the type of microphones, etc. Work presents aproach of independent component analysis in the frequency domain, which can successfully separate also recordings made in the real environment.
|
|
Optimization of UMTS access network
Havlíček, Karel ; Kovář, Petr (referee) ; Novotný, Vít (advisor)
This master’s thesis deals with a UMTS radio access network optimization, containing radio interface analysis, services and procedures description, ways to correct parameters calculations and settings and other issues necessary for a correct access network operation. The goal is effective network operation, minimum costs and maximum performance and flexibility of the network. The optimization during system operation is important because of the system character. The UMTS system uses the WCDMA technology, where particular users share the same frequency band and they are distinguished from one another via code sequences. The capacity of such a system is then given by the interference level - each particular user increases the interference level by the value corresponding to his transmit power. The maximum cell capacity is determined by the maximum interference level at which users can still operate with required services, so it is related not only to the number of users, but also to their bit rate. The optimization allows effective usage of the system for different services with different requirements. The main optimization tool is the radio resource management, containing number of algorithms, such as admission control, which decides the acceptance or rejection of a new user demanding certain service, power control, which ensure that users transmit with a minimal power sufficient for required service, handover and cell selection algorithms etc. The major parameter used by these algorithms is a cell load factor, which is related to the interference level margin. There are several methods for the load factor estimation and some of them are described in this work. Some other optimization techniques are mentioned here, too. This work also contains a laboratory exercise proposal for radio resource management introduction using the OPNET Modeler network simulation tool.
|
|
Luminescence spectroscopy of two-dimensional quantum structures in GaAs/AlGaAs system
Grohoľová, Adela ; Grill, Roman (advisor) ; Oswald, Jiří (referee)
The aim of this work is the study of photoluminescence properties of GaAs/Al0.33Ga0.67As double quantum well. Low-temperature luminescence spectra of this sample are measured in dependence on electric and magnetic eld and dierent excitation power. The temperature dependencies of photoluminescence especially of the indirect excitons in in-plane magnetic eld are gauged as well. The simple model of localized indirect excitons is discussed to explain the discrepancy concerning the damping of indirect exciton photoluminescence in in-plane magnetic eld. The eective g-factors of indirect, neutral and charged excitons are calculated from observed Zeeman splitting. Few simple models are proposed to explain the behavior of eective g-factors. The possible agreement or contradiction with other published experimental data is discussed.
|
| |
guest ::
