|
|
Data Mining Case Study in Python
Stoika, Anastasiia ; Burgetová, Ivana (referee) ; Zendulka, Jaroslav (advisor)
This thesis focuses on basic concepts and techniques of the process known as knowledge discovery from data. The goal is to demonstrate available resources in Python, which enable to perform the steps of this process. The thesis addresses several methods and techniques focused on detection of unusual observations, based on clustering and classification. It discusses data mining task for data with the limited amount of inspection resources. This inspection activity should be used to detect unusual transactions of sales of some company that may indicate fraud attempts by some of its salespeople.
|
|
|
Data Mining with Python
Krestianková, Tamara ; Burgetová, Ivana (referee) ; Zendulka, Jaroslav (advisor)
This thesis deals with principles of data mining process, available Python packages for data mining and a demonstration of Python script capable of data analyisis focused on classification techniques. Created classifiers are able to classify subjects into two groups - healthy people and people suffering from Parkinson's disease - based on their biomedical vocal analysis data.
|
|
|
Social Media Networks in Human Resources Management (Model Supporting a Recruitment of Employees)
Böhmová, Lucie ; Střížová, Vlasta (advisor) ; Basl, Josef (referee) ; Horváthová, Petra (referee)
The main goal and contribution of this thesis are to create an artifact in the form of model that supports a recruitment of employees via social media networks and its implementation into practice. For this purpose has been used a depth analysis of current state of researched problematics leading to finding gaps that could be eventually solved while using mentioned model. The Author takes advantage of knowledge obtained from literature research and also from self-created research, particularly the analysis of content type investigated social media networks and specific data obtained manually and automatically; Comparative analysis of job advertisements; periodic questionnaire survey amongst HR staff. A significant part of the research has also been an investigation of already existing frames of social media networks recruitment. A depth analysis of current state determines that there is existence only of few models dealing with social media networks recruiting. Furthermore, these models lack candidate evaluation based on the social media networks behavior. Artifact as a model that supports a recruitment of employees via social media networks contains a suggestion of an automatic solution dealing with user´s data downloading and also a suggestion of subsequent analytical data processing and the creation of a predictive model for assessing the user´s behavior on the social media networks. Final evaluation of the effectiveness of the proposed model is done through the formal verification process and the case study. The case study verification has the suggested artifact been implemented in practice with a name Model PM that is using a recruiting application (Práce na míru) for an extraction of Facebook´s data. For user´s behavior predictor setting has been used a character test (MBTI). With the help of the cluster analysis and machine learning (Decision trees) has been created Stochastic predictive model that determines a character type of particular candidate (The accuracy of the prediction of the MBTI personality category is in the range between 68 % to 84 % in individual cases with confidence value between 43 % and 81 %). The case study verified a usefulness of model that supports a recruitment of employees via social media networks, and afterward, the mode has been implemented into the practice.
|
|
|
Current trends in customer relationship management
Kotara, Jan ; Jesenský, Daniel (advisor) ; Koudelková, Petra (referee)
The bachelor thesis "Current trends in customer relationship management" deals with - as its title suggests - the issue of CRM and selection of modern factors, which influence its current state. The theoretical part of the thesis presents the definition of customer relationship management and its position in the marketing theory, as well as a brief look into its history. It also contains a section about the specifics of CRM in retail sector, as it is further discussed in the practial part. The explanation of the importance of data and its analysis for CRM is followed by the most important section of the theoretical part about current trends in CRM - especially modern loyalty programs, omnichannel stratégy and social CRM. In the practical part of the thesis these findings are compared with the state of implementation of these trends in the retail chain Tesco. The aim of the thesis is to find the imperfections of the implementation of trends at Tesco and suggest ideas of improvement in order to enhance its customer relationship management.
|
|
|
Application for monitoring and analysis of vascular reconstructions
Drozen, Jan ; Hoksza, David (advisor) ; Škoda, Petr (referee)
Implantation and the care of vascular reconstructions and accesses for haemodialysis is one of areas of cardiovascular surgery. Both vascular reconstructions, known as bypasses, and prosthetic haemodialytic accesses, known as grafts, are used only if conventional therapies are unsuccessful. Thus effectivity of these operations is critically important for patient survival. The aim of this thesis is development of software system for evidence data obtained during vascular operations and following checkups. Important is also support for basic analytical and statistical processing of collected data. Another goal is to transform collected data into knowledge useful for increasing effectivity of vascular operations.
|
|
|
Application for monitoring and analysis of vascular reconstructions
Drozen, Jan ; Hoksza, David (advisor)
Implantation and the care of vascular reconstructions and accesses for haemodialysis is one of areas of cardiovascular surgery. Both vascular reconstructions, known as bypasses, and prosthetic haemodialytic accesses, known as grafts, are used only if conventional therapies are unsuccessful. Thus effectivity of these operations is critically important for patient survival. The aim of this thesis is development of software system for evidence data obtained during vascular operations and following checkups. Important is also support for basic analytical and statistical processing of collected data. Another goal is to transform collected data into knowledge useful for increasing effectivity of vascular operations.
|
|
|
The Use of Data Analytics in Internal and External Audit
Tecáková, Andrea ; Říhová, Zora (advisor) ; Kadochová, Lucie (referee)
Data Analytics is one of the fast-developing applications of IT in organizations worldwide. This Master's thesis examines data analytics in the context of internal and external audit. Principal aim of the thesis is to identify the opportunities for data analytics application in both audit disciplines. Secondary goal is to design a data-analytical procedure, apply it to actual business data and thus demonstrate the benefits of employing data analytics. The thesis builds on a summary of theoretical sources of the relevant area, followed by a survey conducted by the author. The survey maps current state of data analytics usage in both internal and external audit in the Czech Republic. The added value of this thesis is, apart from the identification of audit areas in which it is beneficial to use data analytics, the design of an analytical procedure and its application. Another benefit is the survey revealing current state of the art and the insights of interviewed auditors, pointing to both benefits and problems of data analytics application to the performance of the audit profession.
|
|
|
Applied analytics of Shared service centers
Budka, Marek ; Pour, Jan (advisor) ; Legát, Tomáš (referee)
This thesis focuses on the development and implementation of data analysis in an environment of shared service centers. The aim is the selection of analysis and its implementation from raw data to visualization of the results. The work is divided into two parts. The first part describes the shared services centers, their risks and benefits, and the process of data analysis, including a description and examples of visualization and data quality. The second part of the work describes the development of selected analysis. We chose the analysis of vendor master file. We perform anonymization of production raw data using IBM SPSS Modeler. We are completing this process in Microsoft SQL Server Management Studio, where we are continuing with the development of analysis logic and finally assembling dashboards in Tableau visualization program.
|
|
|
Big Data, their storage and options of exploitation
Macek, Jáchym ; Chlapek, Dušan (advisor) ; Kučera, Jan (referee)
The content of this bachelor's thesis is to analyze work with data especially with large-volume unstructured data, thus Big Data. The thesis is retrieval and contents informational survey based on questionnaires and interviews. The aim is to evaluate and approximate Big Data theme, their storage, tools for their management and opportunities of its exploitation to the reader from technological and business point of view. The objective for the practical part is a survey. The thesis is divided in to three parts. The first part defines the concept, gives closer look at the issue and is dedicated to open and linked data. The second part deals with the question of storage and opportunities of subsequent exploitation of stored data, tools and technologies for Big Data management. Further, there is evaluation of the advantages and disadvantages and comparison of the Hadoop technologies. In the third part the survey of Big Data issue is published. I use tools of questionnaires and interviews for acquiring information from students and experts. Furthermore, it describes current trends, problems and offers solution from the field.
|
|
|
Using software tools to detect fraud
Sobkuliaková, Lucia ; Králíček, Vladimír (advisor) ; Dudová, Lenka (referee)
The subject of this Master thesis is an application of fraud detection softwares. Except computer assisted audit tools and techniques and fraud management systems thesis deal with new methods like using neural network model or logistic regresion model.The thesis also contains comparison of the best softwares for fraud examinations.
|
guest ::
