|
|
Analysis of Data to Solve Problems with Humidity in Buildings
Nečasová, Klára ; Korček, Pavol (referee) ; Kořenek, Jan (advisor)
The aim of this work was to solve problems with excessive humidity in buildings using data analysis. The theoretical part of the work deals with impacts of excessive humidity on the health of building occupants and also the condition of the building structure. Data mining methods including classification, prediction, and clustering are described together with model evaluation and selection. The practical part focuses on hardware platform description and measurement scenarios. Key parameters affecting indoor relative humidity are indoor and outdoor temperature and outdoor relative humidity. The long-term measurement of the mentioned parameters was performed using the set of sensors and BeeeOn system. Measured data was used to design a system for event detection related to a humidity change. The approach to air change regulation in the room was based on natural ventilation.
|
|
|
Speed of sound prediction
Řežábková, Jana ; Hartman, David (advisor) ; Brabec, Marek (referee)
This bachelor thesis presents a novel approach for speed of sound pre- diction in aqueous electrolytic solutions using machine learning techniques. A single model capable of accurately predicting the speed of sound in se- lected electrolytic aqueous solutions at different temperatures and molalities is trained. The machine learning experiment is designed to exploit the dis- sociation of electrolytes in water. Electrolytes are viewed as cation/anion pairs. Therefore, electrolyte description is based purely on its constituting ions. This approach allows to view the available data as a matrix in which rows represent cations, columns anions and each cell a full electrolyte. The idea of being able to fill cells for which no speed of sound data is yet avail- able is tested within the thesis. The final model's accuracy is compared to existent research on speed of sound prediction. However, some of the model approaches are novel and have no existing comparable settings. 1
|
|
|
Systém pro automatický návrh hasičských vozidel
KOTNOUR, Tomáš
The topic of this master´s is development of system for configuration of fire trucks. The design of the fire truck is based on legislative requirements, future incidents and the character of landscape. Processing of current incidents and how to predict future incidents is describe in next chapters. For prediction of the future incidents are used neural networks described the thesis. The main outcome of the thesis is implemented application designed for Windows platform.
|
|
|
Solution of unemployment of young people through closer connestion of their vocational training with labor market requirements
Steindlberger, Martin ; Kotrusová, Miriam (advisor) ; Hiekischová, Michaela (referee)
This diploma thesis deals with the relationship between the vocational training of young people and the requirements of the labour market. The thesis is focused on secondary vocational education. The problem, that has been examined, is the lack of readiness of school graduates, to move to the free labour market. In order to improve this, it is necessary to try to reconcile the content and outputs of vocational training with the needs of the labour market. If graduates are poorly prepared for employers' needs, these graduates may become unemployed. The problem is getting worse with the longer duration of unemployment. The aim of this diploma thesis is to find out and describe who participates in vocational training and how their mutual cooperation is implemented. In the thesis the author will try to find out the strengths and weaknesses of this cooperation. He will also try to find out in what kind of way the training of young people to the needs of the labour market is adapted. The thesis is a case study focused on the South Bohemian Region, which is briefly presented by selected indicators characteristic of the selected region. To obtain the necessary data, there are used figures and statistics from the former researches, as well as information from selected participants involved in the solution...
|
|
|
Předpovídání trendů akciového trhu z novinových článků
Serebryannikova, Anastasia ; Kuboň, Vladislav (advisor) ; Vidová Hladká, Barbora (referee)
In this work we made an attempt to predict the upwards/downwards movement of the S&P 500 index from the news articles published by Bloomberg and Reuters. We employed the SVM classifier and conducted multiple experiments aiming at understanding the shape of the data and the specifics of the task better. As a result, we established the common evaluation settings for all our subsequent experiments. After that we tried incorporating various features into the model and also replicated several approaches previously suggested in the literature. We were able to identify some non-trivial dependencies in the data which helped us achieve a high accuracy on the development set. However, none of the models that we built showed comparable performance on the test set. We have come to the conclusion that whereas some trends or patterns can be identified in a particular dataset, such findings are usually barely transferable to other data. The experiments that we conducted support the idea that the stock market is changing at random and a high quality of prediction may only be achieved on particular sets of data and under very special settings, but not for the task of stock market prediction in general. 1
|
|
|
Artificial neural networks for macroeconomic data analysis
Padrón Peňa, Ildefonso ; Mrázová, Iveta (advisor) ; Kuboň, David (referee)
The analysis and prediction of macroeconomic time-series is a factor of great interest to national policymakers. However, economic analysis and forecast- ing are not simple tasks due to the lack of a precise model for the economy and the influence of external factors, such as weather changes or political decisions. Our research is focused on Spanish speaking countries. In this thesis, we study dif- ferent types of neural networks and their applicability for various analysis tasks, including GDP prediction as well as assessing major trends in the development of the countries. The studied models include multilayered neural networks, recur- sive neural networks, and Kohonen maps. Historical macroeconomic data across 17 Spanish speaking countries, together with France and Germany, over the time period of 1980-2015 is analyzed. This work then compares the performances of various algorithms for training neural networks, and demonstrates the revealed changes in the state of the countries' economies. Further, we provide possible reasons that explain the found trends in the data.
|
|
| |
|
| |
|
|
Web Application for Making Predictions of a Call Centre
Mička, David ; Hynek, Jiří (referee) ; Bartík, Vladimír (advisor)
The goal of this thesis is to create a web application for creation of call centre predictions. The app should be able to replace current solutions that are in use in the daily operation of Kiwi.com s.r.o. The app should be more intuitive and easier to use and maintain than Verint or the spreadsheet solution of doing predictions. It should also have enough options for creation of tactical forecasts that allow the company to react on upcoming situations and should help set realistic expectations for the management of our customer centre.
|
|
|
Prediction of Protein Solubility
Marušiak, Martin ; Martínek, Tomáš (referee) ; Hon, Jiří (advisor)
Protein solubility is closely related to the usability of proteins in industrial use and research. The successful prediction of solubility would therefore lead to a significant saving of financial resources. This work presents new solubility predictor Solpex based on machine learning that achieved better performance on independent test set than any comparable solubility prediction tool. The predictor implementation was preceded by a study of the biological nature of solubility, evaluation of existing solubility prediction approaches, datasets building, many experiments with novel features and selection of the best features for the predictor. As the most important step in machine learning is the datasets building, this work mainly benefits from own rigorous processing of the main source of solubility data - the TargetTrack database.
|
guest ::
