|
|
Deep Neural Networks Used for Customer Support Cases Analysis
Marušic, Marek ; Ryšavý, Ondřej (referee) ; Pluskal, Jan (advisor)
Umelá inteligencia je pozoruhodne populárna v dnešnej dobe, pretože si dokáže poradiť s rôznymi veľmi komplexnými úlohami v odvetviach ako napr. spracovanie obrazu, spracovanie zvuku, spracovanie prirodzeného jazyka a podobne. Keďže Red Hat doteraz už vyriešil obrovksé množstvo zákazníckych požiadavkov počas podpory rôznych produktov. Preto bola navrhnutá myšlienka použiť umelú inteligenciu práve na tieto dáta a docieliť tak zlepšenie a zrýchlenie procesu riešenia zákaznícky požiadavkov. V tejto práci sú popísané použité techniky na spracovanie týchto dát a úlohy, ktoré je možné riešiť pomocou hlbokých neurónových sietí. Taktiež sú v tejto práci popísane rôzne modely, ktoré boli vytvorené počas riešenia tejto práce a snažia sa adresovať rôzne úlohy. Ich výkony sú porovnané na spomínaných úlohách.
|
|
|
Facebook Social Network Datamining and Reconstruction of Captured Communication
Bruckner, Tomáš ; Kmeť, Martin (referee) ; Pluskal, Jan (advisor)
This thesis deals with social network Facebook from perspective of computer forensic science with focus on obtaining sensitive information about tracked users. Its main goal is implementation of the tools for reconstruction of captured communication and Facebook data mining. Core of this application has been implemented in framework for processing of captured communication Netfox.Framework, which is being developed by Faculty of Information Technology, Brno University of Technology. Man-in-the-Middle attack has been used for capturing of decipherable communication. Selenium WebDriver has been used as a tool for data mining. Developed solution is able to reconstruct Facebook conversations between users, addition of new statuses and comments, and interception of sent files. Data mining module is able to obtain public information about tracked users, especially places they recently visited, past and upcoming events, public user details, albums and photos, friendlists and mutual friends with other users. As part of the bachelor thesis, data analysis, implementation of the application, validity testing and benchmark analysis have been performed.
|
|
|
BitTorrent Tracker-Less Peer Monitoring
Vaško, Martin ; Pluskal, Jan (referee) ; Polčák, Libor (advisor)
The goal of this thesis is to effectively monitor peers within BitTorrent network. This research is based on distributed hash tables, specifically MLDHT. In the last decade, focus has also been on efficiency of monitoring within network. Existing algorithms do not provide actual efficiency therefore it is recommended to introduce an estimate accuracy using Bernoulli process. The goal of monitoring is to ensure the highest success of peer search. The work focuses on designing and implementing Bernoulli process model. The measurements show that, the in-depth search (LIFO) is better in the short run. The work focuses on the efficiency of time monitoring, estimation of the peer search error and efficiency using the modeled Bernoulli process. Result of this work is the recommended monitoring time with high efficiency.
|
|
|
Netfox Detective Tool Refactorization
Slámová, Hana ; Polčák, Libor (referee) ; Pluskal, Jan (advisor)
This semestral thesis deals with usage of design patterns in a network tool Netfox Detective. The thesis contains analysis and identification of issues regarding their implementation. Also, a description of writing unit tests for a part of application is present and its automation. This thesis describes Netfox Detective contribution process as well.
|
|
|
User Behavior Anomaly Detection
Petrovič, Lukáš ; Veselý, Vladimír (referee) ; Pluskal, Jan (advisor)
The aim of this work is to create an application that allows modeling of user behavior and subsequent search for anomalies in this behavior. An application entry is a list of actions the user has executed on his workstation. From this information and from information about the events that occurred on this device the behavioral model for a specific time is created. Subsequently, this model is compared to models in different time periods or with other users' models. From this comparison, we can get additional information about user behavior and also detect anomalous behavior. The information about the anomalies is useful to build security software that prevents valuable data from being stolen (from the corporate enviroment).
|
|
|
Distributed Network Traffic Processing
Letavay, Viliam ; Ryšavý, Ondřej (referee) ; Pluskal, Jan (advisor)
Expansion of computer networks and availability of internet connection enables our society to grow faster then ever before. However, at the same time it opens up a new opportunuties for a cybercrime activities. That's why there is an increasing need of security administrators and law enforcing agencies for existence of a tools to analyze the captured data flows. This master thesis deals with ways of analysis of captured network traffic in a distributed environment, which would allow scaling of available analysis power and therefore adapt to ever increasing volumes of data transmitted over the computer networks.
|
|
|
Migration and Refactorization of Netfox Detective for .NET 5
Pokorný, Šimon ; Ryšavý, Ondřej (referee) ; Pluskal, Jan (advisor)
Every second, there are many attempts to attack various entities on the Internet. This is why high-quality, fast, and up-to-date tools are needed to easily analyze network traffic. Netfox Detective is one of such tools. Specifically, it is used for forensic analysis of network communication. The aim of this work is to migrate Netfox Detective to the newest version of .NET platform (.NET 5), including refactoring with respect to user experience and correct use of software design patterns. This thesis deals not only with the migration itself, but is listing common mistakes programmers make along with possible solutions to these mistakes. The chapters contain a detailed decision log that can help guide other developers to better solutions. Furthermore, the work deals with analysis and creation of unit tests and with correct use of tools for CI/CD. Fully migrated project is not the only output of this thesis. A development environment for the project has been prepared in GitLab and it is ready to be used.
|
|
|
DotVVM to Accelerated Mobile Pages Transpiler
Tichý, Michal ; Polčák, Libor (referee) ; Pluskal, Jan (advisor)
This thesis proposes an extension to the DotVVM web framework that adds support for conversion of DotVVM pages into AMP pages. This thesis describes the current status and inner workings of DotVVM and AMP. The thesis contains an overview of the advantages and disadvantages of both technologies. Second part of this thesis describes implementation of DotVVM.AMP. Final part of this thesis shows integration of DotVVM.AMP into an existing DotVVM application.
|
|
|
Room Occupancy Detection with IoT Sensors
Kolarčík, Tomáš ; Jeřábek, Kamil (referee) ; Pluskal, Jan (advisor)
The aim of this work was to create a module for home automation tools Home Assistant. The module is able to determine which room is inhabited and estimate more accurate position of people inside the room. Known GPS location cannot be used for this purpose because it is inaccurate inside buildings and therefore one of the indoor location techniques needs to be used. Solution based on Bluetooth Low Energy wireless technology was chosen. The localization technique is the fingerprinting method, which is based on estimating the position according to the signal strength at any point in space, which are compared with a database of these points using machine learning. The system can be supplemented with motion sensors that ensure a quick response when entering the room. This system can be deployed within a house, apartment or small to medium-sized company to determine the position of people in the building and can serve as a very powerful element of home automation.
|
|
|
.NET's LINQ Optimization
Šerý, Daniel ; Ryšavý, Ondřej (referee) ; Pluskal, Jan (advisor)
This thesis deals with LINQ (Language integrated query) and investigates possibilities of its implementation and optimization in C# language. Method of rewriting of query to procedural code is chosen and implemented. The goal is to provide a LINQ that can be used in code with the need for high speed. Regarding the program created for rewriting LINQ queries, the performance of most operators has been increased by 1.2x to 20x of System.Linq speed depending of rewritten algorithm, data source and provided information to rewriting program.
|
guest ::
RSS feed.