We build a joint multimodal model of text and images for automatically assigning illustrative images to journalistic articles. We approach the task as an unsupervised representation learning problem of finding a common representation that abstracts from individual modalities, inspired by multimodal Deep Boltzmann Machine of Srivastava and Salakhutdinov. We use state-of-the-art image content classification features obtained from the Convolutional Neural Network of Krizhevsky et al. as input "images" and entire documents instead of keywords as input texts. A deep learning and experiment management library Safire has been developed. We have not been able to create a successful retrieval system because of difficulties with training neural networks on the very sparse word observation. However, we have gained substantial understanding of the nature of these difficulties and thus are confident that we will be able to improve in future work.

This thesis deals with the data mining in data stream which represents fast developing area of information technology. The text describes common principles of data mining, explains what data stream is and shows methods for its preprocessing and algorithms for following data mining. The special attention is given to the VFDT and the CVDT algorithm. The next mentioned are the spatiotemporal data and related data mining. The second part describes the design and implementation of the application for classification over spatiotemporal data stream represented by road traffic data and following prediction of spatiotemporal events (traffic-jams). The classification is performed by the VFDT and CVFDT algorithm. The application has been tested on the data set obtained by the simulation tool SUMO.

Cyber-physical systems (CPS) need to be designed to deal with various forms of uncertainty associated with data contributing to the system's knowledge of the environment. Dealing with uncertainty requires adopting an appropriate model, which then allows making the right decisions and carrying out the right actions (possibly affecting the environment) based on imperfect information. However, choosing and incorporating a suitable model into CPS design is difficult, because it requires identifying the kind of uncertainty at hand as well as knowledge of suitable models and their application to dealing with the uncertainty. While inspiration can be found in other CPS designs, the details of dealing with uncertainty in another CPS can be confounded by domain-specific terminology, context, and requirements. To make this aspect of CPS design less daunting, we aim at providing an overview of approaches dealing with uncertainty in the design of CPS targeting collective behavior. To this end, we present a systematic review of relevant scientific projects with industrial leadership and synthesis of relations between system features, the kinds of uncertainty, and methods used to deal with it. The results provide an overview of uncertainty across different domains and challenges and reason about a guide for...

The aim of this thesis is a development of a system for trajectory-based periodic pattern recognition and following GPS trajectory classification. This system is designed according to a performed analysis of techniques of data mining in moving object data and furthermore, on recent research on a subject of a trajectory-based activity recognition. This system is implemented in C++ programming language and experiments addresing its      effectiveness are performed.

The aim of this paper is to present the possibilities of the usage of advanced analytical tools to optimize decision-making in personnel practice. The literature review part of the thesis deals with the so-called HR analytics, its development, possibilities of its usage, and the methodological framework on which it is based. The next part of the paper deals with the specific application of HR analytics in the field of employee retention according to the methodological framework of CRISP-DM. The last chapter describes in detail the phenomenon of employee turnover, its consequences, and possible explanatory variables. The empirical part of the paper is framed as a quantitative, applied research and deals with voluntary turnover of employees in a particular company-a large Czech bank. Firstly, the statistical-inference part of the research identifies several statistically significant predictors of employee turnover through binary logistic regression-unemployment rate, number of changed teams, time spent in the company, salary and total income, salary growth rate, team size, extraordinary bonus, and gender. Secondly, in the data-science part, several prediction models are compiled, one using binary logistic regression as well and another based on several machine learning techniques. The models are...

The forecast of the waste production is an important information for planning in waste management. The historical data often consists of short time series, therefore traditional prognostic approaches fail. The mathematical model for forecasting of future waste production based on spatially distributed data with hierarchically structure is suggested in this thesis. The approach is based on principles of regression analysis with final balance to ensure the compliance of aggregated data values. The selection of the regression function is a part of mathematical model for high-quality description of data trend. In addition, outlier values are cleared, which occur abundantly in the database. The emphasis is on decomposition of extensive model into subtasks, which lead to a simpler implementation. The output of this thesis is tool tested within case study on municipal waste production data in the Czech Republic.

This work deals with issues in data streams mining which nowadays is a very dynamic area in information technology. The thesis describes the general principles of data mining. There are also the principles of data mining in the data streams. Special attention is given to the implemented algorithm CluStream. In the practical part the data stream processing solution was designed and implemented by the MSSQL technology using the above algorithm. The functionality of the algorithm was verified using own data stream generator.

No description available.

The first part of the thesis deals with the methods and tools that can be used to retrieve data from websites and the tools used for data mining. The second part is devoted to practical demonstration of the entire process. Web Czech Dance Sport Federation, which is available on www.csts.cz , is used as the source web site.

The aim of this master´s thesis is to analyse and search unusual dependencies in database of patients from Internal Cardiology Clinic Faculty Hospital Brno. The part of the work is theoretical overview of common data mining methods used in medicine, especially decision trees, naive Bayesian classifier, artificial neural networks and association rules. Looking for unusual dependencies between atributes is realized by association rules and naive Bayesian classifier. The output of this work is a complex system for Knowledge discovery in databases process for any data set. This work was realized with collaboration of Internal Cardiology Clinic Faculty Hospital Brno. All programs were made in Matlab 7.0.1.