|
|
Application of fuzzy logic as a support for decision-making in financial market
Abrahám, Adam ; Šuňavcová, Nikola (referee) ; Janková, Zuzana (advisor)
This thesis focuses on the application of fuzzy logic in supporting of decision-making in financial markets, especially in cryptocurrency investment. Firstly, the theory of fuzzy logic and its various applications are comprehensively discussed. Subsequently, cryptocurrencies and their investment criteria are analyzed. Based on this analysis, a fuzzy system is proposed as a decision-making tool for cryptocurrency investment. Specifically, the fuzzy system enables the evaluation of the risk and return of the investment in a particular cryptocurrency. The implementation of the proposed system, using MS Excel and MathWorks MATLAB, is also described. The experimental results show that the proposed fuzzy system is an effective decision-making tool for cryptocurrency investment.
|
|
|
Fuzzy Models of Driver Behaviour
Fišer, Michal ; Mihálik, Ondrej (referee) ; Jirgl, Miroslav (advisor)
The master's thesis is concerned with finding a suitable fuzzy model for approximating data series representing driver's control interventions during lane changes. Firstly, the theoretical part presents an introduction to driver modelling and human-machine systems, the basics of fuzzy logic, fuzzy systems and controllers, optimization methods and a vehicle driving simulator. The thesis focuses mainly on McRuer models of human operator behaviour in~the area of driver behaviour modelling and genetic algorithms in~the area of optimisation methods. Based on this theory, individual linear models of driver behaviour are subsequently developed. The data were collected on a vehicle driving simulator. On this simulator, several experiments were conducted to measure the driver's response to a step change. Subsequently, the individual driver models were identified from the measured data using the System Identification Toolbox in Matlab. Furthermore, the design of a suitable fuzzy model structure is described. This is followed by~a~description of the genetic algorithm used to identify the fuzzy model and finally the~linear and fuzzy models are compared.
|
|
|
Flat and formalistic approach in law
Pavlíček, Libor ; Maršálek, Pavel (referee)
Flat and formalistic approach in law Many processes in law take place automatically and through inertial force, without admitting one's free will and without encompassing values important to man (freedom, dignity, justice). The bearers of the process of automation in law are Machines (or Automats), i.e. tools ensuring primary legal certainty. However, in addition to the subpage of legal certainty, law also consists of the subpage of justice and effectiveness, which is not controlled by Automats, because their algorithms do not often reflect this subjective dimension in law. The trend of automation in law undoubtedly contributes to a number of improvements, however, there appear risks as well. Since the human soul cannot be programmed by Automats, it cannot be assumed that their decision-making, i.e. their output, will be fair and effective in all circumstances. However, in traditional approach in law this output is considered equal to a binding legal norm. As an example of Automat in the field of mobility the traffic lights may serve, in the field of state administration there are automatic forms or formulas, in the judiciary Automat is represented by a judge acting as a robot (subsumption automat), and artificial intelligence may be seen as the most sophisticated Automat of all. Automats are tools...
|
|
|
Artificial neural networks for clustering and rule extraction
Iša, Jiří
Title: Artificial neural networks for clustering and rule extraction Author: Jiří Iša Department: Department of Software Engineering Supervisor: RNDr. Iveta Mrázová, CSc. Supervisor's e-mail: mrazova@ksi.ms.mff.cuni.cz Keywords: fuzzy, rules, extraction, neural network Abstract: Rule extraction with neural networks has been a com- mon research topic over the last decades. This master thesis proposes a novel growing fuzzy inference neural network, based on the principle of growing neural structures. This allows the network to adjust iteratively its number of hidden neurons. For the purpose of this network an existing clustering algorithm is en- hanced to improve the sensitivity to the requested output. A novel fast weights adaptation, inspired by the fuzzy set theory, is also suggested. The characteristics of the proposed model and a new method of the selection of significant input features support the induction of a relatively small amount of simple fuzzy rules. The introduced techniques have been exper- imentally tested on real-world data describing the relationship between various types of housing in the Boston area and its price. The data was obtained from the "Boston housing" dataset.
|
|
|
Stochastic management storage function of water reservoir using method of artificial intelligence
Kozel, Tomáš ; Fošumpaur, Pavel (referee) ; Zezulák,, Jiří (referee) ; Starý, Miloš (advisor)
The main advantage of stochastic forecasting is fan of possible value, which deterministic method of forecasting could not give us. Future development of random process is described better by stochastic then deterministic forecasting. We can categorize discharge in measurement profile as random process. Stochastic management is worked with dispersion of controlling discharge value. In thesis is described construction and evaluation of adaptive stochastic model base on fuzzy logic, neural networks and evolution algorithm, which are used stochastic forecast from forecasting models described in thesis. The learning fuzzy model and neural network is used as replacement of classic optimization algorithm (evolution algorithm). Model was tested and validated on made up large open water reservoir. Results were evaluated and were compared with model base on traditional algorithms, which was used for 100% forecast (forecasted values are real values). The management of the large open water reservoir with storage function, which was given by stochastic adaptive managing, was logical. The main advantage of fuzzy model and neural network model is computing speed. Classical optimization model is needed much more time for same calculation as fuzzy and neural network model, therefore classic model used clusters for stochastic calculation.
|
|
| |
|
|
Artificial neural networks for clustering and rule extraction
Iša, Jiří
Title: Artificial neural networks for clustering and rule extraction Author: Jiří Iša Department: Department of Software Engineering Supervisor: RNDr. Iveta Mrázová, CSc. Supervisor's e-mail: mrazova@ksi.ms.mff.cuni.cz Keywords: fuzzy, rules, extraction, neural network Abstract: Rule extraction with neural networks has been a com- mon research topic over the last decades. This master thesis proposes a novel growing fuzzy inference neural network, based on the principle of growing neural structures. This allows the network to adjust iteratively its number of hidden neurons. For the purpose of this network an existing clustering algorithm is en- hanced to improve the sensitivity to the requested output. A novel fast weights adaptation, inspired by the fuzzy set theory, is also suggested. The characteristics of the proposed model and a new method of the selection of significant input features support the induction of a relatively small amount of simple fuzzy rules. The introduced techniques have been exper- imentally tested on real-world data describing the relationship between various types of housing in the Boston area and its price. The data was obtained from the "Boston housing" dataset.
|
|
| |
|
|
Autopilot of RC Plane
Denk, Filip ; Bidlo, Michal (referee) ; Mrázek, Vojtěch (advisor)
This thesis is focused on stabilization of the radio-controlled aircraft model. The aim is to create a system that could controll the direction of flight (without pilot intervention) itself. This system also includes determination of the aircraft’s position using a 3-axis gyroscope. The Arduino Uno microcontroller was used as a controll unit.
|
|
|
Tooth Clasification on Jaw 3D Polygonal Model
Hulík, Rostislav ; Španěl, Michal (referee) ; Kršek, Přemysl (advisor)
This document discusses a solution for tooth classification on 3D jaw polygonal model. Sequentially, I describe techniques for representation and browsing of polygonal model saved in computer memory, techniques for dental curve detection and finally, creation of surface representing approximated tooth plane. After it, I analyze possibilities of height map creation from jaw model which helps in tooth classification in the scope of entire dental curve context and, as a last step, final detection of these teeth in two dimensions. In the same time, I discuss 3D polygonal model segmentation for border extraction, which separates teeth from the rest of the model. In the end of proposed algorithm, I join these two runs into one final detection and classification process of separate teeth, so presented application can automatically indentify and classify teeth to corresponding names and positions with a minimum user interaction. In a second half of this document, I describe implemented solution. According to primary goal, I propose these techniques forcefully to multiplatform approach and maximal user comfort.
|
guest ::
