|
|
Software demo of unsupervised learning
Slezák, Milan ; Sáblík, Václav (referee) ; Honzík, Petr (advisor)
The bachelor's thesis introduces the use of unsupervised learning and presents possibilities of cluster analysis. Software demo of unsupervised learning is a part of this thesis. This program was made as a teaching aid. It consists several input databases with different data distributions on the basis of which it is possible to explain very easily elementary principles of cluster analysis and differences between hierarchical clustering and partitional clustering.
|
|
|
Unsupervised learning
Končinský, Petr ; Sáblík, Václav (referee) ; Honzík, Petr (advisor)
The aim of the (diploma) thesis was to identify problems of unsupervised learning. At first I focus on the most essential parts of unsupervised learning. One of the most fundamental parts, on which the project is focused on, is cluster analysis. I give a detailed description of cluster analysis and the final product is a single program which was programmed in C++ language. At analysis I describe tasks with particular variables which I use to run various mathematical operations. At metrics which belong to the most important part of cluster analysis I am concerned with mutual similarity relations between objects and I compute their inter-cluster distances for other clusters. Furthermore, I do not omit other methods belonging to unsupervised learning, principal components analysis and factor analysis are not less important. In my project I attempted to do a survey of methods which are involved in machine learning, in our case in unsupervised learning.
|
|
|
Forex automated trading system based on neural networks
Kačer, Petr ; Honzík, Petr (referee) ; Jirsík, Václav (advisor)
Main goal of this thesis is to create forex automated trading system with possibility to add trading strategies as modules and implementation of trading strategy module based on neural networks. Created trading system is composed of client part for MetaTrader 4 trading platform and server GUI application. Trading strategy modules are implemented as dynamic libraries. Proposed trading strategy uses multilayer neural networks for prediction of direction of 45 minute moving average of close prices in one hour time horizon. Neural networks were able to find relationship between inputs and output and predict drop or growth with success rate higher than 50%. In live demo trading, strategy displayed itself as profitable for currency pair EUR/USD, but it was losing for currency pair GBP/USD. In tests with historical data from year 2014, strategy was profitable for currency pair EUR/USD in case of trading in direction of long-term trend. In case of trading against direction of trend for pair EUR/USD and in case of trading in direction and against direction of trend for pair GBP/USD, strategy was losing.
|
|
|
Effect of HFS Based Feature Selection on Cluster Analysis
Malásek, Jan ; Klusáček, Jan (referee) ; Honzík, Petr (advisor)
Master´s thesis is focused on cluster analysis. Clustering has its roots in many areas, including data mining, statistics, biology and machine learning. The aim of this thesis is to elaborate a recherche of cluster analysis methods, methods for determining number of clusters and a short survey of feature selection methods for unsupervised learning. The very important part of this thesis is software realization for comparing different cluster analysis methods focused on finding optimal number of clusters and sorting data points into correct classes. The program also consists of feature selection HFS method implementation. Experimental methods validation was processed in Matlab environment. The end of master´s thesis compares success of clustering methods using data with known output classes and assesses contribution of feature selection HFS method for unsupervised learning for quality of cluster analysis.
|
|
|
Mobile Based Data Acquisition and Anomaly Detection
Ondrášek, Michael ; Holek, Radovan (referee) ; Honzík, Petr (advisor)
The work deals with the implementation of the specific architecture to detect anomalies in the classroom or in commercial use. The system consists of three parts: Measurement module, mobile applications and server part. Transmission between the measuring module of the server and the evaluation is carried out simultaneously with the visuals on the mobile device. All system components are implemented with the minimum cost and maximum expandability. All the necessary computing power is concentrated in the server part because of usability with multiple simultaneously operating mobile clients. Emphasis is placed on the solution architecture and the possibility of using the system as a whole, or selected portions separately. Finally, experiments are designed for the presentation of selected methods for anomaly detection.
|
|
|
Feature Selection Based on Dynamic Mutual Information
Manga, Marek ; Klusáček, Jan (referee) ; Honzík, Petr (advisor)
This work analyzes and discuss a issue of implementation feature selection method called Dynamic mutual information (DMIFS). Original description of the DMIFS contains several irregularities, therefore DMIFS can not be implemented exactly as original method. Results of implemented DMIFS is compared with results of original DMIFS. This results shows that implemented DMIFS is similar to the DMIFS. Next part of the work describes design of two new methods based on the DMIFS. The first method called DmRMR merges mRMR and DMIFS. Better performance but worse stability of DmRMR was proved by several tests. The second method called WDMIFS is weighted version of the DMIFS based on AdaBoost algorithm. The WDMIFS has worse performance than DMIFS. Finnaly, manual for implementing DMIFS to RapidMiner and Weka is provided.
|
|
| |
|
|
Data Mining
Stehno, David ; Hynčica, Tomáš (referee) ; Honzík, Petr (advisor)
The aim of the thesis was to study and describe data mining methodology CRISP-DM. From the collected database of calls to the call center a prediction was performed, based on CRISP-DM methodology. In phase of test situation modeling four different testing methods were used: the k-NN, neural network, linear regression and super vector machine. The input attributes importance for further prediction was evaluated based on different selections. The results and findings may provide data for further more accurate forecasts in the future; not only in number of calls but also other indicators relevant to the call center.
|
|
|
Ingelligent Import of OSM into the Traffic Simulator TRASI
Muzika, Dávid ; Klusáček, Jan (referee) ; Honzík, Petr (advisor)
The thesis deals with the design and implementation of algorithms for import maps into the simulator TRASI. These algorithms are capable of import map from map portal OpenStreetMaps to the simulation environment. The work deals with adjusting the internal structure of the imported intersections, so that their structure was correct according to the rules of traffic. The work deals with the design and implementation of differential evolution for the design of the structure of intersections.
|
|
|
Design and Testing of Stochastic Navigation in Traffic Simulator TRASI
Erben, Vojtěch ; Drahanský, Martin (referee) ; Honzík, Petr (advisor)
The thesis deals with the design and implementation of routing algorithms in trafic simulator TRASI. These algorithms are capable of planning vehicle's route by giving a set of crossroads that vehicle needs to go through. Furthermore, this work deals with design and implementation of stochastic navigation including implementation of communication between vehicles. Stochastic navigation suggests several alternative routes based on a traffic event. From these routes is randomly (stochastically) choosen one based on information about the throughput of particular found routes. In the introduction of this work is described the traffic simulator TRASI, it's user interface and basic control interface. Further is described theory of traffic flow on macroscopic and microscopic level, followed by the descripion of algorithms for oriented graphs traversal and their implementation in the simulator. In the following parts of this thesis is described communication layer, that takes care of the communication between vehicles, and it's implementation. Further is described design and implementation of stochastic navigation. In the final chapter is done verification of the functionality of the simulator and tests of particular routing algorithms.
|
guest ::
