|
|
Modelling and Analysis of Logistics Processes by Applying Process and Data Mining Techniques
Rudnitckaia, Julia ; Wang, Hao (referee) ; Zendulka, Jaroslav (referee) ; Hruška, Tomáš (advisor)
In this thesis, we propose an approach for modelling hidden and unknown processes and subprocesses in the example of a seaport logistics area. Having the underlying process model makes it possible to exploit more advanced algorithms since deviations and main paths are becoming visible and better controlled. The obtained model is the foundation for the core research of this work and will be enriched with key performing indicators and their forecast by applying advanced process mining, statistics, and machine learning techniques. The main difference of the approach is that we take as a target variable not any specific value, but the object - a process variant or a process type with a set of parameters. Bottleneck analysis, from one side, and predictive analysis, on the other hand, are enforced with context-aware information, especially with these additional objective process attributes. Furthermore, the support of the descriptive ("As is") current process model with certain notation and the integration with relevant bottleneck and predictive methods compromise the advantages of the approach. The work primarily focuses on the design of algorithms and methods for supporting logistics data analysis. However, it can be adjusted and applied to other areas accordingly, which makes the approach flexible and versatile. The result of the work is the framework for unstructured process modelling and the key process parameters predictive method. This analysis of processes with their attributes might be used for decision-making systems and process maps in future.
|
|
|
Algorithm for Product Recommendation
Bodeček, Miroslav ; Bartík, Vladimír (referee) ; Zendulka, Jaroslav (advisor)
The goal of this project is to explore the problem of product recommendations in the area of e-commerce and to evaluate known techniques, design product recommendation system for an existing e-commerce site, implement it and test it. This report introduces the problem, briefly examines current state of affairs in this area and defines requirements for a product recommendation module. The concept of data mining in general is introduced. The report proceeds to present detailed design corresponding to defined requirements and summarizes data gathered during testing phase. It concludes with evaluation and with discussion of the remaining goals for this thesis.
|
|
|
Preemptive Safety Analysis of Road Users' Behavior from Trajectories
Zapletal, Dominik ; Herout, Adam (referee) ; Zendulka, Jaroslav (advisor)
This work deals with the and preemptive road users behaviour safety analysis problem. Safety analysis is based on a processing of road users trajectories obtained from processed aerial videos captured by drons. A system for traffic conflicts detection from spatial-temporal data is presented in this work. The standard approach for pro-active traffic conflict indicators evaluation was extended by simulating traffic objects movement in the scene using Ackerman steering geometry in order to get more accurate results.
|
|
|
SkillMatrix Database
Vorel, Roman ; Burgetová, Ivana (referee) ; Zendulka, Jaroslav (advisor)
This work deals with creation of database information system for employee training and bonus calculation with direct connection on existing personal systems in Honeywell production company. The project makes provision for solution on dynamic database server and maximal usage of its potentiality for application development. The system dispose of access privileges containing several user roles and role of administrator. The system concept is realized in UML modeling language. The implementation is realized on Microsoft technologies. The system is generally implmented in 3 tiers: user interface, database component, data base model. The user interface is implemented in ASP.NET 2.0 with utilization of ASP.NET AJAX 1.0 technology. The database component is implemented using object oriented language Visual .NET C# and ADO.NET classes. Data base is relized on SQL Server 2000 in Transact-SQL language. The integral part of the project is also testing on sample data and application in live business on Honeywells company intranet.
|
|
|
Configuration and Reservation System for Concert Halls
Sikora, Vít ; Burget, Radek (referee) ; Zendulka, Jaroslav (advisor)
This paper describes the implementation of a web application that enables configuring events in concert halls and defining sections and seats in those halls. Furthermore, application can run in a mode for end users that enables placing reservations in previously defined halls. Certain emphasis is placed on customizability so that it would be easy to integrate this application into an existing system. Implementation is divided into server part (REST API) realized using PHP 7 with Restler framework and to client part built as a single page application using modern Javascript (ECMAScript 2016) with React framework and transpiled to common Javascript and HTML using Webpack and Babel.
|
|
|
Classification on unbalanced data
Hlosta, Martin ; Popelínský, Lubomír (referee) ; Štěpánková,, Olga (referee) ; Zendulka, Jaroslav (advisor)
Tématem této disertační práce je klasifikace daty s nevyváženými daty. Jedná se o oblast strojového, jejímž cílem je řešit problémy, které plynou z toho, že jedna ze tříd je v datech zastoupena výrazně méně než třída druhá. Minoritní třída má často větší význam a tradiční metody upřednostňující majoritní třídu nedosahují dobrých výsledků na třídě minoritní. Dvě aplikační domény motivovaly výzkum a vedly na identifikaci dvou specifických, dosud neřešených problémů. V první z nich vedlo omezení kladené na minimální požadovanou přesnost na minoritní třídě v počítačové bezpečnosti na formulaci úlohy klasifikace s omezením. Navrhl jsem metodu, která kombinuje upravenou verzi logistické regrese a stochastické algoritmy, které vždy vylepšily výsledky logistické regrese.Druhou je doména analýzy učení (Learning Analytics), která motivovala definici problému predikce splnění cíle, jenž má specifikovaný termín splnění. Byl představen koncept sebe-učení (Self-Learning), kdy trénování modelu probíhá díky jedincům, kteří tento cíl splní předčasně. Díky malému počtu jedinců splňujících úlohu na začátku je problém silně nevyvážený, ale nevyváženost klesá směrem k termínu splnění. Na problému identifikace rizikových studentů distanční univerzity bylo ukázáno, že (1) takový koncept dává lepší výsledky než specifikovaná základna (baseline), (2) a že metody pro vypořádání se s nevyvážeností, které neberou v potaz informaci o doméně, nevedly k velkým zlepšením. Evaluace ukázala, že metody založené na znalosti domény v rozšířené verzi pro Self-Learning vylepšily klasifikaci více než běžné metody pro vypořádání se s nevyvážeností a že znalost příčiny nevyváženosti může vést k lepším výsledkům.
|
|
|
Tool for Payment Format Definition Debugging
Kuba, Richard ; Rychlý, Marek (referee) ; Zendulka, Jaroslav (advisor)
The main goal of this thesis is to develop and demonstrate a tool for debugging payment formats that would make it easier for users of the DMEEX transaction (program) to detect bugs in definition trees. Demonstration of the debugging tool is implemented on the SAP S / 4HANA platform. The first part of this thesis describes the platforms SAP R / 3 and SAP S / 4HANA, with emphasis on capturing the differences between them. Furthermore, the purpose of payment formats and their integration within SAP systems is discussed. The design describes the collection and work with user requirements for this tool. The implemented product allows users to visualize the processing of the DMEEX transaction definition tree, thanks to which it allows its users to more easily detect errors in the definition of the tree or in its input data.
|
|
|
Multi-Level Association Rules Module of a Data Mining System
Pospíšil, Jan ; Bartík, Vladimír (referee) ; Zendulka, Jaroslav (advisor)
This thesis focuses on the problematics of implementing a multilevel association rules mining module, for existing data mining project. There are two main algorithms explained, Apriori and MLT2L1. The thesis continues with the datamining module implementation, as well as the DMSL elements design. In the final chapters deal with an example dataminig task and its result comparison as well as the whole thesis achievement description.
|
|
|
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.
|
|
|
Web Application of Recommender System
Koníček, Igor ; Bartík, Vladimír (referee) ; Zendulka, Jaroslav (advisor)
This master's thesis describes creation of recommender system that is used in real server cbdb.cz. A~fully operational recommender system was developed using collaborative and content-based filtering techniques. Thanks to many user feedback, we were able to evaluate their opinion. Many recommended books were tagged as desirable. This thesis is extending current functionality of cbdb.cz with recommender system. This system uses its extensive database of ratings, users and books.
|
guest ::
