|
|
Utilization of artificial intelligence in technical diagnostics
Konečný, Antonín ; Huzlík, Rostislav (referee) ; Zuth, Daniel (advisor)
The diploma thesis is focused on the use of artificial intelligence methods for evaluating the fault condition of machinery. The evaluated data are from a vibrodiagnostic model for simulation of static and dynamic unbalances. The machine learning methods are applied, specifically supervised learning. The thesis describes the Spyder software environment, its alternatives, and the Python programming language, in which the scripts are written. It contains an overview with a description of the libraries (Scikit-learn, SciPy, Pandas ...) and methods — K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Decision Trees (DT) and Random Forests Classifiers (RF). The results of the classification are visualized in the confusion matrix for each method. The appendix includes written scripts for feature engineering, hyperparameter tuning, evaluation of learning success and classification with visualization of the result.
|
|
|
Localization of People in the Building
Randýsek, Vojtěch ; Orság, Filip (referee) ; Sakin, Martin (advisor)
The aim of this thesis is to create a personal indoor positioning system. In this work, new systembasedonaBluetoothtechnologyisproposed.Personlocalizationisaccomplishedby usingthefingerprintingmethod.Proposedsystemconsistsofadatabase,awebapplication, a hardware layer and communication services. For the purpose of demonstration of the proposed system a video was made. System in the video, using propsed methods, gained standard accuracy of 2.46 m and standard deviation of 2 m. These results show that the combination of used technology leads to a succesful localization of people in buildings.
|
|
|
Instance based learning
Martikán, Miroslav ; Polách, Petr (referee) ; Honzík, Petr (advisor)
This thesis is specialized in instance based learning algorithms. Main goal is to create an application for educational purposes. There are instance based learning algorithms (IBL), nearest neighbor algorithms and kd-trees described theoretically in this thesis. Practical part is about making of tutorial application. Application can generate data, classified them with nearest neighbor algorithm and is able of IB1, IB2 and IB3 algorithm testing.
|
|
|
Routing Problems and Their Solutions
Pospíšil, Václav ; Dvořák, Jiří (referee) ; Šeda, Miloš (advisor)
The first part of the thesis is devoted to an Introduction and a comprehensive description of all important concepts of graph theory, which is followed by descriptions and modifications of two selected types of routing problems: the travelling salesman problem and the vehicle routing problem. The next part of the thesis deals with subsequent possibilities of solving problems through deterministic and stochastic algorithms. It also includes a practical part, which at the end of the thesis deals with the shortest path optimization of the two created models using Nearest neighbour algorithm, Genetic algorithm and solver in GAMS.
|
|
|
Detection of atrial fibrillation in ECG
Húsková, Michaela ; Vítek, Martin (referee) ; Maršánová, Lucie (advisor)
Aim of this thesis is description of problems of atrial fibrillation and methods that could be used for detection in the electrocardiogram. The introductory part of the theoretical analysis deals with the principle of electrophysiology of the heart and mainly the pathophysiology of atrial fibrillation. Additionally the work is focused on describing methods on automatic atrial fibrillation detection and capabilities of PhysioNet database. In the practical part methods are implemented in the MATLAB environment. After using the statistics to evaluate the quality of the parameters, the automatic classification of the data was performed by the method of The Nearest Neighbour. Finally, the accuracy of testing is presented.
|
|
|
Routing Problems and Their Solutions
Pospíšil, Václav ; Dvořák, Jiří (referee) ; Šeda, Miloš (advisor)
The first part of the thesis is devoted to an Introduction and a comprehensive description of all important concepts of graph theory, which is followed by descriptions and modifications of two selected types of routing problems: the travelling salesman problem and the vehicle routing problem. The next part of the thesis deals with subsequent possibilities of solving problems through deterministic and stochastic algorithms. It also includes a practical part, which at the end of the thesis deals with the shortest path optimization of the two created models using Nearest neighbour algorithm, Genetic algorithm and solver in GAMS.
|
|
|
Utilization of artificial intelligence in technical diagnostics
Konečný, Antonín ; Huzlík, Rostislav (referee) ; Zuth, Daniel (advisor)
The diploma thesis is focused on the use of artificial intelligence methods for evaluating the fault condition of machinery. The evaluated data are from a vibrodiagnostic model for simulation of static and dynamic unbalances. The machine learning methods are applied, specifically supervised learning. The thesis describes the Spyder software environment, its alternatives, and the Python programming language, in which the scripts are written. It contains an overview with a description of the libraries (Scikit-learn, SciPy, Pandas ...) and methods — K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Decision Trees (DT) and Random Forests Classifiers (RF). The results of the classification are visualized in the confusion matrix for each method. The appendix includes written scripts for feature engineering, hyperparameter tuning, evaluation of learning success and classification with visualization of the result.
|
|
|
Divergent standards of living among similar households
Charvát, Daniel
The diploma thesis examines sources of poverty and material deprivation among households that exhibit similar characteristics in the Czech Republic. The data originate from the European Union - Survey on Income and Living Conditions between years 2009 - 2015. For the analysis, two rounds of binomial logistic regressions were carried out with poverty status and material deprivation as dependent factors. First, a total of 28 400 unique households were included in both models. Afterwards, a nearest-neighbor matching procedure was used to pair households that (1) live in poverty and (2) are materially deprived. That led to (1) 5 378 and (2) 9 406 observations for the subsequent estimations. The outcomes suggest that even when the similarity between households is accounted for the unemployment status is the most influential factor of poverty and material deprivation, alongside with low equivalised disposable income. Furthermore, the households for which housing costs entail great burden are among the more threatened. Numerous households are more likely to be poor and deprived while larger houses and flats are related with lesser odds of poverty and material deprivation.
|
|
|
Localization of People in the Building
Randýsek, Vojtěch ; Orság, Filip (referee) ; Sakin, Martin (advisor)
The aim of this thesis is to create a personal indoor positioning system. In this work, new systembasedonaBluetoothtechnologyisproposed.Personlocalizationisaccomplishedby usingthefingerprintingmethod.Proposedsystemconsistsofadatabase,awebapplication, a hardware layer and communication services. For the purpose of demonstration of the proposed system a video was made. System in the video, using propsed methods, gained standard accuracy of 2.46 m and standard deviation of 2 m. These results show that the combination of used technology leads to a succesful localization of people in buildings.
|
|
|
Detection of atrial fibrillation in ECG
Húsková, Michaela ; Vítek, Martin (referee) ; Maršánová, Lucie (advisor)
Aim of this thesis is description of problems of atrial fibrillation and methods that could be used for detection in the electrocardiogram. The introductory part of the theoretical analysis deals with the principle of electrophysiology of the heart and mainly the pathophysiology of atrial fibrillation. Additionally the work is focused on describing methods on automatic atrial fibrillation detection and capabilities of PhysioNet database. In the practical part methods are implemented in the MATLAB environment. After using the statistics to evaluate the quality of the parameters, the automatic classification of the data was performed by the method of The Nearest Neighbour. Finally, the accuracy of testing is presented.
|
guest ::
