|
|
Galaxy Generator for Games
Březina, Karel ; Kobrtek, Jozef (referee) ; Starka, Tomáš (advisor)
This thesis is focused on design and implementation of procedural generator of galaxy for games or simulations. Second goal is implementation of demonstration application which is showing possible usage of generator. Generator is able to create galaxy by predefined types or by user created map.
|
|
|
Memory Reduction of Stateful Network Traffic Processing
Hlaváček, Martin ; Puš, Viktor (referee) ; Kořenek, Jan (advisor)
This master thesis deals with the problems of memory reduction in the stateful network traffic processing. Its goal is to explore new possibilities of memory reduction during network processing. As an introduction this thesis provides motivation and reasons for need to search new method for the memory reduction. In the following part there are theoretical analyses of NetFlow technology and two basic methods which can in principle reduce memory demands of stateful processing. Later on, there is described the design and implementation of solution which contains the application of these two methods to NetFlow architecture. The final part of this work summarizes the main properties of this solution during interaction with real data.
|
|
|
Cluster analysis of data traffic
Nagyová, Simona ; Martinásek, Zdeněk (referee) ; Blažek, Petr (advisor)
This thesis is meant to present the methods of cluster analysis, their use for detecting network traffic anomalies. It deals with various types of cyber attacks particularly DoS (DDoS) attacks. It contains a list of well-known DoS attacks. Also includes a program in Python, using the k-means method for detection of anomalous traffic.
|
|
|
Success Rate Measure Methods in Data Mining
Trunkát, Jan ; Zelený, Jan (referee) ; Bartík, Vladimír (advisor)
The Bachelor thesis is aimed at success rate measure methods in data mining in the area of clustering. It introduces the basic concepts, features of data mining and especially the cluster analysis. This work includes program, which implements methods of measuring success. In conclusion, they are given results of clustering success.
|
|
|
Hledání shluků v grafech
Navrátil, Jan ; Lidický, Bernard (advisor) ; Mareš, Martin (referee)
The goal of this thesis is to create an application which will be able to identify clusters in graphs. The application contains modified algorithms from cluster analysis and graph algorithms used for identifying communities in complex networks. The purpose of this work is not speed optimalisation of implementation but the opportunity to try and compare results of each algorithm and verify whether special graph algorithms are truly better. The core of this task was to modify algorithms from classical cluster analysis (four were chosen, with a set of settings each) to work with graphs and implement and present them within one application along with community detection algorithms (two, in more versions each).
|
|
|
Galaxy Generator for Games
Březina, Karel ; Kobrtek, Jozef (referee) ; Starka, Tomáš (advisor)
This thesis is focused on design and implementation of procedural generator of galaxy for games or simulations. Second goal is implementation of demonstration application which is showing possible usage of generator. Generator is able to create galaxy by predefined types or by user created map.
|
|
|
Hledání shluků v grafech
Navrátil, Jan ; Lidický, Bernard (advisor) ; Mareš, Martin (referee)
The goal of this thesis is to create an application which will be able to identify clusters in graphs. The application contains modified algorithms from cluster analysis and graph algorithms used for identifying communities in complex networks. The purpose of this work is not speed optimalisation of implementation but the opportunity to try and compare results of each algorithm and verify whether special graph algorithms are truly better. The core of this task was to modify algorithms from classical cluster analysis (four were chosen, with a set of settings each) to work with graphs and implement and present them within one application along with community detection algorithms (two, in more versions each).
|
|
|
Statistical analysis of selected wage level indicators in regions of the Czech Republic
Čálková, Tereza ; Kába, Bohumil (advisor) ; Jindřich, Jindřich (referee)
The thesis describes the development of selected wage indicators in regions of the Czech Republic in the period 2002-2014 and predicts the future development of these indicators for the next three years. Subsequently thesis classifies individual regions into clusters based on similarity in terms of wage indicators and selected labour indicators and then compares these clusters within two years - 2002 and 2014. Foundations are taken from the website of the Ministry of Labour and Social Affairs and Public database of Czech Statistical Office. The thesis is divided into theoretical and practical part and contains eight chapters. The theoretical part explains terms of remuneration of workers, describes the selected indicators, regions and used methods. In the practical part is applied time series analysis, principal component analysis and cluster analysis. The results of the analyses are summarized in the conclusion of the thesis.
|
|
|
Cluster analysis of data traffic
Nagyová, Simona ; Martinásek, Zdeněk (referee) ; Blažek, Petr (advisor)
This thesis is meant to present the methods of cluster analysis, their use for detecting network traffic anomalies. It deals with various types of cyber attacks particularly DoS (DDoS) attacks. It contains a list of well-known DoS attacks. Also includes a program in Python, using the k-means method for detection of anomalous traffic.
|
|
|
Success Rate Measure Methods in Data Mining
Trunkát, Jan ; Zelený, Jan (referee) ; Bartík, Vladimír (advisor)
The Bachelor thesis is aimed at success rate measure methods in data mining in the area of clustering. It introduces the basic concepts, features of data mining and especially the cluster analysis. This work includes program, which implements methods of measuring success. In conclusion, they are given results of clustering success.
|
guest ::
