|
|
Automation of data preprocessing using domain knowledge
Beskyba, Jan ; Šimůnek, Milan (advisor) ; Pejčoch, David (referee)
In this work we propose a solution that would help automate the part of knowledge discovery in databases. Domain knowledge has an important role in the automation process which is necessary to include into the proposed program for data preparation. In the introduction to this work, we focus on the theoretical basis of knowledge discovery of databases with an emphasis on domain knowledge. Next, we focus on the basic principles of data pre-processing and scripting language LMCL that could be part of the design of the newly established applications for automated data preparation. Subsequently, we will deal with application design for data pre-processing, which will be verified on the data the House of Commons.
|
|
|
Automation of a data mining process by the LISp-Miner system
Ochodnická, Zuzana ; Šimůnek, Milan (advisor) ; Rauch, Jan (referee)
This thesis is focused on the area of automated data mining. The aim of this thesis is a description of the area of automated data mining, creation of a design of an automated data mining tasks creation process for verification of set domain knowledge and new knowledge search, and also an implementation of verification of set domain knowledge of attribute dependency type influence with search space adjustments. The implementation language is the LMCL language that enables usage of the LISp-Miner system's functionality in an automated way. These data analyses were performed on data from air pollution monitoring. The design and implementation were successful and the created scripts could be used (with some manual changes in initial parameters) for analyses of another dataset as well.
|
|
|
Means of Processing of Documents in DocBook stored in ZIP Archive
Hrachovec, Jakub ; Kosek, Jiří (advisor) ; Šimůnek, Milan (referee)
The purpose of this thesis is to define a format based on ZIP archive for DocBook documents which would allow their easier handling, e.g. copying or sending them as an attachment with an e-mail. The second goal is creation of an appliaction in the Java programming language which would allow packaging of DocBook documents into the defined format and subsequently their processing and transformation into PDF, HTML and another formats. The results of this thesis allow easier sharing of DocBook documents which are packaged into an archive alongside files referenced from these documents. Firstly, I am going to describe two already existing formats that are based on ZIP archive and these formats are going to inspire me during the definition of a suitable format for DocBook documents that include images and stylesheets. Afterwards I am going to describe operation and funcionality of the created application and finally I am going to prepare a DocBook document which will be packaged into a DocBook archive and transformed into PDF and HTML.
|
|
|
Maturity model of source code of object-oriented aplications
Pavlíčková, Jarmila ; Novotný, Ota (advisor) ; Šimůnek, Milan (referee) ; Havlíček, Zdeněk (referee)
The goal of this disertation was to develop a maturity model for source code of object-oriented applications and to use this model to verify the quality of the source code of student's applications. The starting point of the thesis was devoted to the analysis of existing standards, norms, methodologies and summaries of best practice for assessing the quality of software products and analysis of the factors that affect the quality of the source code of object applications. To validate the results of the analysis of these factors, the analysis was complemented with a field research conducted among specialist with programming experience. The model to determine the maturity of source code of object-oriented applications was designed according to the analysis and the questionnaire. The statistical method of cluster analysis was used in the design of the model. The model was designed and the procedure of its use in the evaluation of source code was described. This model was pilot tested in education program at the University of Economics in Prague.
|
|
|
Repository for results of association rules data mining tasks in SEWEBAR project
Marek, Tomáš ; Šimůnek, Milan (advisor) ; Svátek, Vojtěch (referee)
This diploma thesis aims at design and implementation of I:ZI Repository application. I:ZI Repository application provides management of data mining tasks and theirs results repository and functions for search in this repository. I:ZI Repository is a REST API build on top of Java EE technology, Berkeley XML database is used for storing data mining tasks. I:ZI Repository application was created based on XQuery search application. The application has completely new structure compared to XQuery search application, all functionality of XQuery search application is present in I:ZI Repository application. Possibilities of using more general search query was added into I:ZI Repository application as well as fuzzy approaches for searching and possibility of clustering search results. Enhanced logging of application activities aimed at logging incoming search queries and outgoing search results is a part of implementation. Results of application testing are included as well.
|
|
|
Generating data using the LM Reverse-Miner
Stluka, Jakub ; Šimůnek, Milan (advisor) ; Kliegr, Tomáš (referee)
In past years, great attention has been paid to evolutionary algorithms and they have been utilized in wide range of industries including data mining field, which nowadays presents a highly demanded product for many commercial institutions. Both mentioned topics are combined in this work. Main thesis subject is testing of new Reverse-Miner module, which can generate data with hidden properties using evolutionary algorithms while using also other modules of LISp-Miner system, commonly used for the purposes of data mining. Main goal lies in generation of two databases by the module in such way so they would meet explicitly set requirements. Other goals are also set within the thesis in the form of understanding the domain necessary for subsequent modeling. The result of the practical part of the thesis is represented not only by two successfully generated databases, but also by description of steps, methods and techniques used. The common recommendations for data preparation by module Reverse-Miner are later summarized, based on experience with modeling. Previous thesis outputs are furthermore contemplating the conclusion of analysis of technical means used for generation and they also provide several suggestions for possible future extensions.
|
|
|
Data mining v oblasti Customer relationship managemet: Analýza dat významné logistické společnosti
Nekvapil, Viktor ; Rauch, Jan (advisor) ; Šimůnek, Milan (referee)
The thesis addresses possibilities of deploying the open source data mining system LISp-Miner in the customer relationship management (CRM), specifically in the area of lead management. This is basically a process of finding information about potential customers, qualifying those customers according to their potential (future value), and turning the selected potential customers to real customers. The data used includes the records concerning the lead management of a major logistic company operating worldwide (the company wanted to stay in anonymity). The data is analysed using the LISp-Miner system which is an academic software developed at the Faculty of Informatics and Statistics at the University of Economics, Prague. The thesis also pays attention to the collaboration with the business experts of the company which provided the data. The principle aim of the thesis is to provide information contributing to the possible change of internal processes of the company. Further aims are to propose directions of the use of the LISp-Miner system when solving a similar data mining task, and propose a simple and understandable way how to present the results. The aims have been achieved by doing the analysis in compliance with the CRISP-DM Methodology. The asset of the thesis is the description of the whole project which includes the analysis of real data using the LISp-Miner system. Further result of the thesis is the description of the lead management domain. Finally, the thesis offers the instructions and recommendations for future similar projects. Section I outlines the LISp-Miner system and its procedures. Section II - A case study describes the process of analysing the data. Two cycles ("iterations") of the analysis were performed. The chapters devoted to the both iterations have been structured according to the phases of the CRISP-DM Methodology. Section III summarises the observations gained during the entire project. Moreover, it gives recommendations and instructions for the application in case a similar project of the data analysis using the LISp-Miner system is designed.
|
|
|
The Analysis of Real CRM Data by the LISp-Miner System
Ochodnická, Zuzana ; Šimůnek, Milan (advisor) ; Rauch, Jan (referee)
This bachelor thesis is focused on in-depth analysis -- data mining of real CRM data. The analysis will be proceeded by CF-Miner and KL-Miner procedures of LISp-Miner system. The aim is to use these procedures on real CRM data, which may lead to discovering various relations among the analysed data. In order to reach this aim, I will use the CRIPS-DM methodology, which is a data mining methodology describing the whole process of data analysis. My contribution will be in the description of CF-Miner and KL-Miner procedures usage which could help other students and people work with these procedures and my contribution will also be in the data analysis which could lead to better understanding of the data so that they can be used more effectively.
|
|
|
The use of genetic programming in forex trading strategy development
Líbal, Petr ; Šimůnek, Milan (advisor) ; Berka, Petr (referee)
This thesis deals with trading strategy development on foreign market exchange (FOREX) from a perspective of technical analysis. A hypothesis that price data can be used for stable profitable decision making is analyzed. For that purpose, financial data preparation and derived indicators are described in detail. At first, strategies are randomly constructed. Afterward, they are improved iteratively by means of evolution principles. Genetic programming is used in particular. Special attention is devoted to fitness functon definition, on which the progress of strategies depends. Besides usual criterion - profit, other aspects are taken into account. Transactional costs are included in strategy evaluation. Costs have even been increased in comparison with declared values. Some of the best strategies are tested on data, which has not been used for their development. Obtained results did not support the hypothesis. However, process of a strategy search can be improved.
|
|
|
Visualisation of musical notation using XML technologies
Vacek, Matěj ; Kosek, Jiří (advisor) ; Šimůnek, Milan (referee)
This thesis focuses on the visualization of musical notation. Even for such a small group of data, there is a specification (MusicXML) which can be used for converting between different softwares. The development of musical notation has been going on for centuries and it has become very complex by now, because composers want to express all their thoughts. Therefore, I have decided to limit the supported part of musical notation and in this thesis, I will focus only on the area of simple Czech folk songs, which have their own specifics. The songs will be displayed in SVG format. This format is based on vector graphics. Transfer between the original and SVG file will be implemented by using XSLT transformation, which is created in this thesis. As far as I am concerned, there is no such already existing transformation.
|
guest ::
RSS feed.