|
Sequential Pattern Mining
Tisoň, Zdeněk ; Zendulka, Jaroslav (referee) ; Hlosta, Martin (advisor)
This master's thesis is focused on knowledge discovery from databases, especially on methods of mining sequential patterns. Individual methods of mining sequential patterns are described in detail. Further, this work deals with extending the platform Microsoft SQL Server Analysis Services of new mining algorithms. In the practical part of this thesis, plugins for mining sequential patterns are implemented into MS SQL Server. In the last part, these algorithms are compared on different data sets.
|
|
Advanced retinal vessel segmentation methods in colour fundus images
Svoboda, Ondřej ; Jan, Jiří (referee) ; Odstrčilík, Jan (advisor)
Segmentation of vasculature tree is an important step of the process of image processing. There are many methods of automatic blood vessel segmentation. These methods are based on matched filters, pattern recognition or image classification. Use of automatic retinal image processing greatly simplifies and accelerates retinal images diagnosis. The aim of the automatic image segmentation algorithms is thresholding. This work primarily deals with retinal image thresholding. We discuss a few works using local and global image thresholding and supervised image classification to segmentation of blood tree from retinal images. Subsequently is to set of results from two different methods used image classification and discuss effectiveness of the vessel segmentation. Use image classification instead of global thresholding changed statistics of first method on healthy part of HRF. Sensitivity and accuracy decreased to 62,32 %, respectively 94,99 %. Specificity increased to 95,75 %. Second method achieved sensitivity 69.24 %, specificity 98.86% and 95.29 % accuracy. Combining the results of both methods achieved sensitivity up to72.48%, specificity to 98.59% and the accuracy to 95.75%. This confirmed the assumption that the classifier will achieve better results. At the same time, was shown that extend the feature vector combining the results from both methods have increased sensitivity, specificity and accuracy.
|
|
Romany Housing
Vyhlídalová, Lucie ; Manová, Ludmila (referee) ; Menšíková, Naděžda (advisor) ; Šuhajda, Karel (advisor)
The project of Romany Housing is located in Brno, the urban district Maloměřice. Architecturally, it is designed as a low-rise row housing, taking into consideration the special requirements of the Romani community, such as living in larger families. Functionally, the structure is designed as a socialized housing with certain level of government support. The housing development is variable. The basis of the project are two types of houses, which alternate in the row-house structure. Moreover, each one of these types also offers different solutions altering the capacity. The houses are designed to be easily extended, even while being used. The construction is wooden, using environmentally-friendly material for cladding – ecopanels. The rows of houses follow the perimeter of the parallelogram of the site and create a peaceful relaxation area in the middle, suitable for free time activities. From the architectural point of view, the new housing development fits into the current image of the location in this quiet part of Brno, which is dominated by small houses with gabled roofs. The new buildings are characterized by asymmetrical gable and windows playfully placed in an asymmetrical pattern. Another distinctive element is the facade, which combines trapezoidal metal sheet and wooden cladding.
|
|
Trends of the consumer behaviour during the global economic recession: Detailed image of the Czech consumer
Kadilák, Michal ; Chytková, Zuzana (advisor) ; Stříteský, Václav (referee)
The main goal of this bachelor thesis is to explain the perfumes consumption trend in the Czech economic background and its differentiation from other, mostly savings-oriented trends. I am interested not only to describe the main signs of the trend, but I will also focus my research on unveiling hidden motivators inside consumer himself/herself. To fulfill this assignment, I chose the methods of qualitative market research to gather and interpret relevant data. Research itself was conducted with 5 consumers using individual in-depth interviews. During interviews, I was consecutively realizing the differences between general consumption pattern during the crisis and the consumption of perfumes, as one of the factors of self-presentation. By the end of the research, all the main drivers of consumers spending towards this specific category were identified. All findings were compared and connected to the theories of well known experts in the field of customer behaviour study, Cooley's "looking-glass self" and Belk's -- "extended self". The perfumes consumption trend confirmed its existence in the Czech Republic among respondents and has shown its tight correlation to consumers' identity and personality.
|
|
Snižování rizik u transakcí na trhu s marihuanou v České republice a v USA - instituce přátelství
Běláčková, Vendula ; Ježek, Tomáš (advisor) ; Zábranský, Tomáš (referee) ; Dušek, Libor (referee)
BACKGROUND: Economists depict illicit markets as violent, due to the lack of centralized property rights enforcement. At the same time, the importance of friendship networks and drug sharing is a recently documented feature of the marijuana market. Recent studies show an increased role of acquiring marijuana through friends, especially in settings where drug policy is rather punitive. This thesis extends this research into the norms that marijuana users attribute to their definition friendship. To do this, the thesis conceptualizes friendship as a type of institution that reduces the transaction costs on the market, and like that, it limits the decision making of marijuana market players. DATA: Marijuana market patterns in the Czech Republic and North-Central Florida were analyzed via both qualitative and quantitative research methods. For the purpose of the qualitative study, 44 (resp 66) study participants were marijuana users and retailers recruited at North-Central Florida (resp in the Czech Republic), with the use of respondent-driven sampling. Inclusion criteria into the study was the use of marijuana in the last 12 months. Semi-structured interviews, that took 80 minutes on average, followed an interview guide focused on marijuana use, sharing, purchases, sales and growing, with extensive probes on activities of respondents` "friends", as they defined them. As for the quantitative data, marijuana market modules from two representative general population surveys on substance use were used (CS 2008, NSDUH). METHODS: Qualitative data were analysed with the use of inductive analysis, and were framed into institutional economics theory. Quantitative data were analyzed with the use ordinary logit models. FINDINGS: The study has shown remarkable impact of drug policies on cannabis markets via comparison between the Czech Republic and the U. S. (North-Central Florida). The study findings suggest that users' definitions of friendship include expectations for behavior that sustain the distribution chain within the marijuana markets. Respondents provided definitions of friendship that contained norms on marijuana sharing and reciprocation, purchases for friends, and introduction to the dealer - for whom the term "friend" has been used as a synonym in most cases. In quantitative analysis, acquistion through a friend made significant reduction of price at last purchase in the U. S., approaving the hypothesis that friendship can be an effective institution to reduce transaction costs on the market. In the Czech Republic, such analysis was inconclusive. This demonstrates that the importance of friendship might be higher in countries where drug prohibition is more severe. CONCLUSIONS: Punitive drug policy provides incentives to shrinking the market into social networks, and like that, it imposes harms on users in terms of decreasing control over their substance use can criminal risks (larger amounts purchased, and the risk of detection to regular citizens, who serve as middlemen on the market without an intention to make profit). For more precise estimates, further surveys shall distinguish between different modalities of friendship, and between different product types.
|
|
Procedure SDKL-Miner for data mining
Tegze, Miron ; Lín, Václav (referee) ; Rauch, Jan (advisor)
A new data mining GUHA procedure SDKL-Miner is presented which extends current data mining system LISp-Miner developed on the VŠE Praha. The procedure mines for various patterns based on evaluation of two two-dimensional contingency tables. A simple comparing quantifier is implemented. Tests and new results for this quantifier are presented together with set of examples of it's use. New directions of future development are proposed. The use of statistical dependency tests and the use of both vector and matrix majorization are analyzed in more details. The procedure is a result of continued development of the academic LISp-Miner system for knowledge discovery and data mining.
|
|
Spin dynamics in GaAs-based semiconductor structures
Schmoranzerová, Eva ; Němec, Petr (advisor) ; Postava, Kamil (referee) ; Kužel, Petr (referee)
This work is dedicated to the study of spin dynamics in systems based on the semiconductor gallium arsenide (GaAs) that are suitable for use in spintronic devices. We explored two types of model structures using experimental methods of ultrafast laser spectroscopy and transport measurements. In the ferromagnetic semiconductor (Ga,Mn)As, we investigated laser-induced magnetization precession. We found out that transfer of both energy and angular momentum from the circularly polarized laser light can trigger magnetization precession, the latter one being identified as a new phenomenon, the "optical spin transfer torque". Furthermore, we demonstrate the possibility to control the energy-transfer-induced magnetization dynamics both optically and electrically using piezo-stressing. When dealing with purely non-magnetic structures for spintronics, we studied the Spin-Injection Hall Effect (SIHE) in GaAs/AlGaAs heterostructures with a special type of spin- orbit (SO) coupling that are lithographically patterned to create nanodevices. We managed to observe precession of the electron spin in the SO field directly in the space domain by extending the original detection method. This finding, together with the direct detection of a pure spin current, helped to propose a working spin Hall effect transistor.
|
|
Nuptiality in Kazakhstan during the period of rapid economic growth
Kozhakhmetova, Aliya ; Kocourková, Jiřina (referee) ; Rychtaříková, Jitka (advisor)
This study primarily addresses nuptiality patterns in Kazakhstan over the period 1999-2007 characterized with rapid economic growth. The data are taken from censuses, vital statistics and other relevant data sources. The data collection is provided by the Agency of Statistics of the Republic of Kazakhstan and the UN Statistics Division. The aim of the thesis is to study marital behavior in the Republic of Kazakhstan, in particular, on differentials by sex, age, nationality and place of residence. During the studied period nuptiality rates arose significantly. Therefore, the question is to what extend the favorable economic development had the impact on positive changes in marital behavior of population. Keywords: Nuptiality, Marriage, Divorce, Kazakhstan, USSR, Economic Growth
|
|
Combining temporal logic and behavior protocols
Mandys, Petr ; Kofroň, Jan (referee) ; Adámek, Jiří (advisor)
In this thesis we consider one of the weaknesses of temporal logic - the fact that the temporal formulas specifying complex properties are hard to read. We introduce new temporal logic "BP-CTL", that originate from Computational Tree Logic (CTL) extended with operators partly taken from Behavior Protocols (BP) and partly newly defined. Text of the thesis is divided into several parts. First we introduce reader to the context of the issue. Next we describe new operators and show their usage on small examples. Then we formally define the resulting language (BP-CTL). In the next part we demonstrate the usability of BP-CTL and introduce the tool - called bpctl - for checking properties written in BP-CTL. Finally we evaluate and conclude our work. The text is extended with appendixes including detailed description of used formalisms, mapping tables of patterns collected in Property Specification Patterns project for BP-CTL and bpctl user manual.
|
|
Structural Colors of Self-Similar Nano Patterns
Meluzín, Petr ; Horáček, Miroslav ; Krátký, Stanislav ; Kolařík, Vladimír
E-beam lithography is a flexible technology for various diffraction gratings origination. The e-beam patterning typically allows for the creation of optical diffraction gratings in the first diffraction order. However, the very high resolution enables also the patterning of structures providing the zero-order diffraction. Recently, we presented a work on the structural colors of metallic layers covering both regular-line structures and CGH (computer generated hologram) structures. This work presents a study dealing with zero-order diffraction structures with self-similar properties. The practical part of the work is focused on two aspects: design parameters and technological issues. Variations in design parameters include the tone of the structure (positive or negative),\nthe density of filling, the filling factor, and the depth of the structures. The achieved gamut of colors may by primarily extended by the proper selection of metal deposition technology and its parameters, and further by the proper selection of the metal and the thickness of its layer, these are the technological issues.
|
guest ::
