|
|
Use Machine Learning to Predict Future Market Prices
Klhůfek, Michal ; Trchalík, Roman (referee) ; Holkovič, Martin (advisor)
This thesis discusses a market prediction system based on the data obtained from the historic tranzaction. The main goal was to use the techniques of technical analysis to create a more accurate estimation of market behavior in the future. The data obtained from the current state of the market are compared with the historical market values using the algorithms for the classification of data from the field of learning. Based on individual algorithms, the software was designed to try to match the two sets of data as closely as possible. Testing took place on a dataset that represented the past market enthusiasm, and how much the overall system is performing.
|
|
|
Segmetation of tomographic data in 3D Slicer
Korčuška, Robert ; Dvořák, Pavel (referee) ; Mikulka, Jan (advisor)
This thesis contains basic theoretical information about SVM-based image segmentation and data classification. Basic information about 3D Slicer software are presented. Aspects of medical images segmentation are described. Workplan and implemetation of SVM method for MRI segmentation in 3D Slicer sofware as extension module is created. SVM method is compared with simple segmentation algorithms included in 3D Slicer. Quality of segmentation, based on SVM, tested on real subjects is experimentaly demonstrated.
|
|
|
Recognition of electrochemical signals using artificial neuronal network
Šílený, Jan ; Kuchta, Radek (referee) ; Hubálek, Jaromír (advisor)
Automatical electrochemical measurements are sources of large data sets intended for further analysis. This work deals with classification, evaluation and processing of electrochemical signals using artificial neural networks. Due to high dimensionality of input data, an autoassociative neural network (AANN) is used in this work. This type of network performs dimensionality reduction via filtering the input data into relatively small number of principal parameters at the bottleneck output. These extracted parameters can be used for classification, evaluation and additional modelling of analyzed data trough the reconstructive part of this network. Furthermore, this work deals with implementation of a feedforward neural network in OpenCL language.
|
|
|
Intelligent Client for Music Player Daemon
Wagner, Tomáš ; Kočí, Radek (referee) ; Janoušek, Vladimír (advisor)
The content of this master thesis project is about design and implementation of intelligent client application for Music Player Daemon (MPD), which searches and presents the metadata related to played content. The actual design precedes the theoretical analysis, which includes analysis of agent systems, methods of data classification, web communication protocols and languages for describing HTML document. At the same time is analyzed the MPD server and communication protocol used by clients application. Furthermore, this work describes the current client applications that presents metadata. In the last chapters of the thesis describes the design and implementation of intelligent client. It describes the methods of solution the implementation and solution of problems. Lastest chapters describes the testing result.
|
|
|
Response analysis of train track laboratory model
Heteš, Marek ; Věchet, Stanislav (referee) ; Kšica, Filip (advisor)
In terms of safety, railway tracks have to be kept in good condition. Early and accurate detection of track defects can save both time and money. This thesis deals with simulation of a passing train on laboratory apparatus, and the measurement of the generated response. Apparatus represents scaled down section of a railway track, and allows the simulation of defect formation. With the help of the obtained data, a suitable method for defect detection was created.
|
|
|
Data Classification using Artificial Neural Networks
Gurecká, Hana ; Dvořák, Jiří (referee) ; Matoušek, Radomil (advisor)
The thesis deals with neural networks used in data classification. The theoretical part presents the three basic types of neural networks used in data classification. These networks are feedforward neural network with backpropagation algorithm, the Hopfield network with minimization of energy function and the Kohonen’s method of self-organizing maps. In the second part of the thesis these algorithms are programmed and tested in Matlab environment. At the end of each network testing results are discussed.
|
|
|
Response analysis of train track laboratory model
Heteš, Marek ; Věchet, Stanislav (referee) ; Kšica, Filip (advisor)
In terms of safety, railway tracks have to be kept in good condition. Early and accurate detection of track defects can save both time and money. This thesis deals with simulation of a passing train on laboratory apparatus, and the measurement of the generated response. Apparatus represents scaled down section of a railway track, and allows the simulation of defect formation. With the help of the obtained data, a suitable method for defect detection was created.
|
|
|
The Depth of Functional Data.
Nagy, Stanislav ; Hlubinka, Daniel (advisor) ; Omelka, Marek (referee)
The depth function (functional) is a modern nonparametric statistical analysis tool for (finite-dimensional) data with lots of practical applications. In the present work we focus on the possibilities of the extension of the depth concept onto a functional data case. In the case of finite-dimensional functional data the isomorphism between the functional space and the finite-dimensional Euclidean space will be utilized in order to introduce the induced functional data depths. A theorem about induced depths' properties will be proven and on several examples the possibilities and restraints of it's practical applications will be shown. Moreover, we describe and demonstrate the advantages and disadvantages of the established depth functionals used in the literature (Fraiman-Muniz depths and band depths). In order to facilitate the outcoming drawbacks of known depths, we propose new, K-band depth based on the inference extension from continuous to smooth functions. Several important properties of the K-band depth will be derived. On a final supervised classification simulation study the reasonability of practical use of the new approach will be shown. As a conclusion, the computational complexity of all presented depth functionals will be compared.
|
|
|
Matlab Implementation Of Multilayer Perceptron For Bearing Faults Classification
Doseděl, Martin
This paper deals with implementation of multilayer perceptron neural network (NN) forbearing faults classification. Neural network has been created from scratch as an M-script with backpropagation learning algorithm also, but without using advanced MATLAB packages. Public availablebearing dataset from CaseWestern Reserve University has been used for both training and testingphase, as well as for the final classification process. Problem with sparse input data for training thenetwork has also been addressed. This relatively simple and small neural network is capable to classifythe failures of a bearing with very low error rate.
|
|
|
Use Machine Learning to Predict Future Market Prices
Klhůfek, Michal ; Trchalík, Roman (referee) ; Holkovič, Martin (advisor)
This thesis discusses a market prediction system based on the data obtained from the historic tranzaction. The main goal was to use the techniques of technical analysis to create a more accurate estimation of market behavior in the future. The data obtained from the current state of the market are compared with the historical market values using the algorithms for the classification of data from the field of learning. Based on individual algorithms, the software was designed to try to match the two sets of data as closely as possible. Testing took place on a dataset that represented the past market enthusiasm, and how much the overall system is performing.
|
guest ::
