Thesis--Münster.

This thesis is focused on using data mining in Business Intelligence solution. The goal of this thesis is implementation of data mining in Business Intelligence solution. Integration of these technologies is posible to obtain synergic effect. The thesis is made in context of academic project Farfalia and results are used in the project. The goal of project is to create component for a new form of reporting and that is using data mining as a tool for data analysis. The benefit of the implementation is automatic browsing of multidimensional data cube / multidimensional data model to find interesting KPI trends, interesting patterns or anomalies in the data. Tasks of data mining are realized by analytic models that are created in this thesis. These models are prepared in R language. Developed scripts of models are universal and after their adjustment by constraints of specific project are useable in any BI solution.

Data clustering plays an important role in effective analysis of gene expression. Although DNA microarray technology facilitates expression monitoring, several challenges arise when dealing with gene expression datasets. Some of these challenges are the enormous number of genes, the dimensionality of the data, and the change of data over time. The genetic groups which are biologically interlinked can be identified through clustering. This project aims to clarify the steps to apply clustering analysis of genes involved in a published dataset. The methodology for this project includes the selection of the dataset representation, the selection of gene datasets, Similarity Matrix Selection, the selection of clustering algorithm, and analysis tool. R language with the focus of Kmeans, fpc, hclust, and heatmap3 packages in R is used in this project as an analysis tool. Different clustering algorithms are used on Spellman dataset to illustrate how genes are grouped together in clusters which help to understand our genetic behaviors.

This thesis aims to describe the theoretical principles of customer segmentation using modern methods of cluster analysis and their practical verification in a commercial environment of utility company. The theoretical part of the thesis describes the principles of clustering analysis using K-means algorithm and hierarchical clustering algorithms. The SW support chapter deals with the possibilities of R, Rapid Miner and KXEN software tools. For the purpose of data processing the theoretical part describes SEMMA, 5A and CRISP methodologies. The practical part is focused on the analysis of homogeneity/heterogeneity of the energy market in the household segment in the utility company EON. Using data from company´s customer system, available external data and programming language R a hierarchical cluster analysis is processed. Through the analysis of gained clusters there are identified the main characteristics of particular clusters which are consequently examined from marketing usage perspectives.

nÃo hà / The main objective of this work is to present a site evaluation study for the Power performance generator of small scale wind turbines (AEPP). The Brazilian norm ABNT NBR IEC 61400-12-1 and its annexes A (obstacle valuation) and B (terrain assessment) were used as reference. Two source codes were written in the R language in order to automate the evaluation process. Both source codes were validated through direct comparison between the results of the developed R codes and the results obtained by DNV GL, which has ISO17025 international laboratory certification for power performance measurements of wind turbines according to the IEC standard. The validated code was then used to evaluate possible positions for the AEPP power performance measurement in UFC. The topographic map was downloaded from the free database SRTM and the characteristic obstacles were determined by satellite images and photos of the area. Altogether, three areas and their respective pairs AEPP â anemometric tower were analyzed regarding their available measurement sector and the necessity or not to perform site calibration. / Este trabalho tem o objetivo de apresentar o estudo de avaliaÃÃo do local para mediÃÃo de curva de potÃncia de aerogeradores de pequeno porte (AEPP). Foi utilizada como referÃncia a norma brasileira ABNT NBR IEC61400-12-1 e seus anexos A (avaliaÃÃo de obstÃculos) e B (avaliaÃÃo do terreno). Foram escritos dois cÃdigos-fonte na linguagem R para automatizar o processo de avaliaÃÃo. Os cÃdigos-fonte foram validados por comparaÃÃo direta entre os resultados obtidos pelos cÃdigos desenvolvidos em R e os resultados obtidos pela empresa DNV GL, que possui certificaÃÃo internacional de laboratÃrio ISO17025 para mediÃÃo de curva de potÃncia de aerogeradores segundo a norma IEC. O cÃdigo validado foi entÃo utilizado para avaliar possÃveis locais para mediÃÃo de curva de potÃncia de AEPP na prÃpria UFC. O mapa topogrÃfico do local foi adquirido da base de dados gratuita SRTM e os obstÃculos do local foram caracterizados por imagens de satÃlite e fotos da regiÃo. No total, trÃs Ãreas e os respectivos pares AEPP â torre anemomÃtrica foram analisados em relaÃÃo ao setor de mediÃÃo disponÃvel e a necessidade ou nÃo de calibraÃÃo do local.

This thesis deals with usage of data mining on data of a library. It specializes in data of users and their behavior registered by library management system Clavius. For this purpose the data of University Library of Silesian University in Opava were used. Firstly the data mining therm is explained and it's use in libraries is thoroughly described. Secondly the ways of usage of the Clavius data are discussed then the text follows data mining usage for management of the library in particular. Finally the results of data mining concerning user's behavior is described in the rear of this thesis. Powered by TCPDF (www.tcpdf.org)