|
Data Mining for Suggesting Further Actions
Veselovský, Martin ; Burget, Radek (referee) ; Bartík, Vladimír (advisor)
Knowledge discovery from databases is a complex issue involving integration, data preparation, data mining using machine learning methods and visualization of results. The thesis deals with the whole process of knowledge discovery, especially with the issue of data warehousing, where it offers the design and implementation of a specific data warehouse for the company ROI Hunter, a.s. In the field of data mining, the work focuses on the classification and forecasting of the advertising data available from the prepared data warehouse and, in particular, on the decision tree classification. When predicting the development of new ads, emphasis is put on the rationale for the prediction as well as the proposal to adjust the ad settings so that the prediction ends positively and, with a certain likelihood, the ads actually get better results.
|
|
Simple Recommender System
Gorčák, Damián ; Rychlý, Marek (referee) ; Bartík, Vladimír (advisor)
Recommender systems are very important in searching for items all over the internet. There are many algorithms for creating recommendations. The main goal of this thesis was to find suitable datasets and make application, which would process them. After that, chosen algorithms for recommender systems are compared with selected datasets
|
|
Travel time prediction
Mudroch, Andrej ; Janáková, Ilona (referee) ; Honec, Peter (advisor)
This thesis discusses travel time prediction of vehicles on roads based on the methods of machine learning. It describes theory of travel times and summarizes scientific papers dealing with this topic. Within the thesis, analysis of real travel time data was done and the features to be used in prediction models were engineered. Finally, the complex prediction system was designed and implemented and has been tested in production environment.
|
|
Very Low Bit-Rate Speech Coding Based on Neural Networks
Jochman, Stanislav ; Malenovský, Vladimír (referee) ; Černocký, Jan (advisor)
Vrámci tejto práce sme skúmali možnosti zlepšenia kvality zvuku produkovaným pomocou neurónovej siete LPCNet. Analyzovali sme vplyv použitia dátových setov zameraných na cieľový jazyk a ich vplyv na kvalitu výsledného zvuku. Pre meranie kvality kódovania reči sme využili hodnotiaci systém WARP-Q. Cieľom našej práce bolo navrhnúť vylepšenie trénovacieho dátového setu a použitie postfilterov pre zlepšenie kvality zvuku. Naše výsledky ukazujú merateľné zlepšenia s využitím malého slovenského dátového setu. Rovnako sme zaznamenali, že využitie dolnopriepustného filteru a filtra zlepšujúceho formanty zlepšilo kvalitu výsledného zvuku.
|
|
Simple Recommender System
Gorčák, Damián ; Rychlý, Marek (referee) ; Bartík, Vladimír (advisor)
Recommender systems are very important in searching for items all over the internet. There are many algorithms for creating recommendations. The main goal of this thesis was to find suitable datasets and make application, which would process them. After that, chosen algorithms for recommender systems are compared with selected datasets
|
|
Optimization of Run Configurations of k-Wave Jobs
Sasák, Tomáš ; Jaroš, Marta (referee) ; Jaroš, Jiří (advisor)
This thesis focuses on scheduling, i.e. correct approximation of configurations used to run k-Wave simulations on supercomputers from the IT4Innovations infrastructure. Especially, for clusters Salomon and Anselm. A single work is composed of a set which contains many simulations. Every simulation is executed by some code from the k-Wave toolbox. To calculate the simulation, it is necesarry to select a suitable configuration, which means the amount of supercomputer resources (number of nodes, i.e. cores), and the duration of the rental. Creation of an ideal configuration is complicated and is even harder for an inexperienced user. The approximation is made based on the empiric data, obtained from multiple executions of different sets of simulations on given clusters. This data is stored and used by a set of approximators, which performs the actual approximation by methods of interpolation and regression. The text describes the implementation of the final scheduler. By experimenting, the most efficient methods for this problem has found out to be Akima spline, PCHIP interpolation and cubic spline. The main contribution of this work is creation of a tool which can find suitable configuration for k-Wave simulation without knowing the code or having lots of experience with its usage.
|
|
Air Quality Analysis in Office and Residential Areas
Tisovčík, Peter ; Korček, Pavol (referee) ; Kořenek, Jan (advisor)
The goal of the thesis was to study the indoor air quality measurement focusing on the concentration of carbon dioxide. Within the theoretical part, data mining including basic classification methods and approaches to dimensionality reduction was introduced. In addition, the principles of the developed system within IoTCloud project and available possibilities for measurement of necessary quantities were studied. In the practical part, the suitable sensors for given rooms were selected and long-term measurement was performed. Measured data was used to create the system for window opening detection and for the design of appropriate way of air change regulation in a room. The aim of regulation was to improve air quality using natural ventilation.
|
|
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.
|
|
Machine Learning Optimization of KPI Prediction
Haris, Daniel ; Burget, Radek (referee) ; Bartík, Vladimír (advisor)
This thesis aims to optimize the machine learning algorithms for predicting KPI metrics for an organization. The organization is predicting whether projects meet planned deadlines of the last phase of development process using machine learning. The work focuses on the analysis of prediction models and sets the goal of selecting new candidate models for the prediction system. We have implemented a system that automatically selects the best feature variables for learning. Trained models were evaluated by several performance metrics and the best candidates were chosen for the prediction. Candidate models achieved higher accuracy, which means, that the prediction system provides more reliable responses. We suggested other improvements that could increase the accuracy of the forecast.
|
|
The Use of Artificial Intelligence for Decision Making
Nezbedová, Katarína ; Pekárek, Jan (referee) ; Dostál, Petr (advisor)
This bachelor thesis deals with the Tamari attractor problem and its application for forming a prediction model. The core of the work is to create a simulation program in the MATLAB development environment and to use it to create and compare several case studies of a predictive model based on different parameters. This model is graphically illustrated and supplemented by economic interpretation.
|