|
|
Modelling of logistics of inter-municipal waste transport
Kučera, Jiří ; Popela, Pavel (referee) ; Nevrlý, Vlastimír (advisor)
This diploma thesis focuses on the issue of inter-municipal waste transport. The main part of the thesis focuses on the development of a computing core designed to create a collection plan for selected municipalities forming a union. The computing core was created so that it could be integrated into a user-friendly web application. The creation of a computational tool is based on the theoretical foundations of graph theory, mathematical programming and heuristic algorithms. In this thesis, consecutive algorithms for clustering of address points, collection design for several types of waste with a heterogeneous vehicle fleet and for creating collection proposal for a several-week long collection plan were developed. The presented approach was tested on real data from an existing union of municipalities in the South Moravian Region.
|
|
|
Implementation of KDE+ statistic method
Svoboda, Tomáš ; Popela, Pavel (referee) ; Bíl,, Michal (advisor)
In this master's thesis I presented a new statistical method KDE+ (Kernel Density Estimation plus) that allows detecting clusters of points on the linear data. I created a self-standing application that enables anybody to try the method and apply it on their own data. One possible usage of the method and application is for the detection of critical roads sections with a high concentration of traffic accidents. Development of the application includes analysis of KDE+ statistical method, design of appropriate program structures and the implementation. Optimization were carried out to achieve higher performance after creating the prototype. At the end the software was validated by analysing vehicle collision data from the police database of the Czech Republic.
|
|
|
HOUSE ON THE EDGE – Valašské Meziříčí, Sokolská Street
Pomykalová, Andrea ; Vojtová, Lea (referee) ; Hrabec, Josef (advisor)
The architectural and urban concept is based on the original plan distribution plots (long and narrow) and conservation way for pedestrians (from the square to the neighborhood). The actual shape of the building is the philosophy of two masses showing the connection of two cities (Meziříčí and Krasno) in one city. The connection of these masses in response from two rivers in the city. (Vsetínská and Rožnovská), which is most visible from the street Sokol. Because of noisy road on the south side of the parcel has been beneficial to the orientation of buildings on the east and west which is also consistent with binding to the original parcel. Two symmetrical objects continue from roofs on existing buildings in the street Pospisilova.
|
|
|
Cluster Analysis Module of a Data Mining System
Riedl, Pavel ; Burgetová, Ivana (referee) ; Zendulka, Jaroslav (advisor)
This master's thesis deals with development of a module for a data mining system, which is being developed on FIT. The first part describes the general knowledge discovery process and cluster analysis including cluster validation; it also describes Oracle Data Mining including algorithms, which it uses for clustering. At the end it deals with the system and the technologies it uses, such as NetBeans Platform and DMSL. The second part describes design of a clustering module and a module used to compare its results. It also deals with visualization of cluster analysis results and shows the achievements.
|
|
|
The Application of PSO in Business
Veselý, Filip ; Kaštovský, Petr (referee) ; Dostál, Petr (advisor)
This work deals with two optimization problems, traveling salesman problem and cluster analysis. Solution of these optimization problems are applied on INVEA-TECH company needs. It shortly describes questions of optimization and some optimization techniques. Closely deals with swarm intelligence, strictly speaking particle swarm intelligence. Part of this work is recherché of variants of particle swarm optimization algorithm. The second part describes PSO algorithms solving clustering problem and traveling salesman problem and their implementation in Matlab language.
|
|
|
ECG arrhythmia detection
Šoltés, Tomáš ; Filipenská, Marina (referee) ; Novotná, Petra (advisor)
This bachelor thesis describes commonly present arrhytmias such as premature ventricular complex, bundle branch blocks and their detection using conventional methods and modern methods, utilising neural networks. Practical part includes: Detection of premature ventricular complexes by their RR intervals and detection of bundle branch blocks using statistical analysis of QRS complex and their clustering with K-nearest neighbors
|
|
|
Evolutionary Optimization of Convolutional Neural Networks
Čoupek, Vojtěch ; Mrázek, Vojtěch (referee) ; Sekanina, Lukáš (advisor)
This thesis deals with the problem of neural network weights compression using the technique of Weight-Sharing and parameter optimization of this technique by unconventional optimization algorithms. The reason for the optimization is decreasing the memory or energy demands of the neural network response calculation. The aim is to design a system that accepts a neural network and reduces its memory demands. Its functionality is demonstrated with the help of several experiments. The thesis investigates the use of various optimization algorithms, additional compression using the quantization above the Weight-Sharing technique, and proposes the quantization results tuning method to improve accuracy. These procedures are first tested on the Le-Net-5 network and then applied for the MobileNet\_v2. network compression.
|
|
|
Image-Based Clustering of Microbial Colonies
Láncoš, Jan ; Kišš, Martin (referee) ; Beneš, Karel (advisor)
Laboratorní analýza mikrobiálních kolonií pěstovaných na Petriho miskách je v současné době předmětem intenzivního výzkumu ve snaze o dosažení totální laboratorní automatizace. Jádro tohoto problému spočívá v přesné lokalizaci kolonií při obrazové analýze. Současná řešení nejčastěji využívají strojové učení, které je však obecně závislé na kvalitních datech, kterých je v tomto odvětví k dispozici jen velmi málo. Abych adresoval tento problém, vytvořil jsem víceúčelový generátor syntetických dat. Úspěšně jsem jej aplikoval jak při segmentaci tak při shlukování kolonií. Výsledné segmentační F1 skóre se mi podařilo navýšit z 0,518 na 0,729 a při shlukování jsem s využitím této segmentace dosáhl V-measure skóre 0,830. Prací navržený přístup ke generování syntetických dat nás posouvá o krok blíže k plné laboratorní automatizaci.
|
|
|
Data Sets for Network Security
Setinský, Jiří ; Hranický, Radek (referee) ; Tisovčík, Peter (advisor)
In network security, machine learning techniques are used to effectively detect anomalies and malware in network traffic. A quality dataset is needed to train a network classifier with high accuracy. The aim of this paper is to modify the dataset using machine learning techniques to improve the quality of the dataset which will lead to training the model with a higher accuracy. The dataset is analyzed by a clustering algorithm and each cluster is characterized by a statistical description resulting from the attributes of the input dataset. The statistical description along with the information of the original classifier is used to compute the score. The score serves as a weight in the modification phase. Cluster analysis allows to filter out the data that are important for training the final model. The proposed approach allows us to mitigate the redundancy of the dataset or to augment it with missing data. The result is a modification framework that is able to reduce the datasets or perform their aggregation in order to create a compact dataset that reflects the actual network traffic. Models were trained on the created datasets and achieved higher accuracy compared to the existing solution.
|
|
|
Visualizing Neural Network Used as a Language Model
Ryšánek, Jakub ; Černocký, Jan (referee) ; Beneš, Karel (advisor)
LSTM sít je typ neuronové sítě, která je určena na analýzu sekvenčních dat. Výhodou LSTM oproti jednoduché rekurentní neuronové síti je schopnost ukládat dlouhodobé závis- losti, což umožňuje dosahovat vyšší úspěšnosti při provádění úloh jako je rozpoznávání řeči nebo jazykové modelování. Avšak vzhledem z jejich komplexitě není zcela jasné jak přesně fungují. Abych prozkoumal jejich vnitřní chování tak jsem vytvořil tři vizualizační metody. Tyto metody se zaměřují na vzor chování jednotlivých prvků modelu nebo na chování celého modelu při zpracování slov s podobným syntaktickým nebo sémantickým významem.
|
guest ::
