|
|
Social Network Analysis using methods of pattern recognition
Križan, Viliam ; Burget, Radim (referee) ; Atassi, Hicham (advisor)
Diplomová práca sa zaoberá rozpoznávaním emócií z textu v sociálnych sieťach. Práca popisuje súčasné metódy extrakcie príznakov, používané lexikóny, korpusy a klasifikátory. Emócie boli rozpoznávané na základe klasifikátoru, netrénovaného na anotovaných dátach z mikroblogovacej siete Twitter. Výhodou použitia služby Twitter, bolo geografické vymedzenie dát, ktoré umožňuje sledovanie zmien emócií populácie v rôznych mestách. Prvým prístupom klasifikácie bolo vytvorenie Baseline algoritmu, ktorý používal jednoduchý lexikón. Pre zlepšenie klasifikácie sme v druhom bode použili komplexnejší SVM klasifikátor. SVM klasifikátory, extrakcie a selekcie príznakov boli použité z dostupnej Python knižnice Scikit. Dáta pre natrénovanie klasifikátoru boli zhromažďované z oblasti USA, a to s pomocou vytvorenej aplikácie. Klasifikátor bol natrénovaný na dátach, označených pri ich zhromažďovaní - bez manuálnej anotácie. Boli použité dve rôzne implantácie SVM klasifikátorov. Výsledné klasifikované emócie, v rôznych mestách a dňoch, boli zobrazené v podobe farebných značiek na mape.
|
|
|
Authentication using a mobile phone
Križan, Viliam ; Hajný, Jan (referee) ; Malina, Lukáš (advisor)
Bachelor’s thesis describes the basics of authentication methods by knowledge, used authentication tokens and biometric ways of authentication. It discusses the capabilities of the individual methods designed for mobile phones. It covers the basic security in the GSM network (using SIM cards) and usage of authentication based on passwords, PIN codes, gestures, facial recognition and speech recognition. Thesis evaluates individual methods in terms of success and performance. It discusses the usage possibilities of authentication methods on the Android platform, where it provides the options of modular arithmetic operations of cryptographic primitives. The work evaluates the time consumption of modular arithmetic operations on the mobile phones with created application. It also lists available algorithms of symmetric and asymmetric ciphers, one-way hashes on the Android platform. The section of authentication protocol conception writes about protocol based on zero-knowledge proof, which authenticates users anonymously. It also proposes the transfer of calculated authentication variables using NFC technology. The protocol also supports the revocation of malicious users. The goal of the thesis was to create two applications that have implemented the selected authentication protocol. The first application designed for users, runs on mobile phones with Android operating system. Application calculates the authentication variables using authentication parameter which it loads from secure element of the phone. After calculations it sends the authentication variables via NFC reader (connection phone – reader). NFC reader is connected to the second application, which verifies the user’s authentication variables and decides the conclusion of authentication. The last part of thesis is devoted to testing the developed applications and running them on the devices. It evaluates the security of the authentication protocol, usage of the phone’s secure element and a security of the transmission over NFC.
|
|
|
Processes Supporting Decision-Making
Križan, Viliam ; Uher, Václav (referee) ; Karásek, Jan (advisor)
This thesis deals with algorithms for supporting decision processes. Firstly, Analytic Hierachy Process (AHP) and Analytic Network Process (ANP) developed by prof. Thomas L. Saaty are described. Basic principles and the implementation in Java programing language of the decision processes are explained. Both processes are then analyzed and their pros and cons and also practical aspect are explained. Secondly, the genetic algorithm NSGA-II developed by Kalyanmoy Deb is described. There are also basic principles and implementation in Java programing language explained. There are finaly the results of NSAG-II algorithm presented.
|
|
|
Random number generator
Križan, Viliam ; Burda, Karel (referee) ; Ležák, Petr (advisor)
This master thesis deals with a generation of random numbers and Fortuna generator implementation in Java language. In the first part the theoretical familiarization to the issues is introduced. Various entropy sources like mouse movement, keyboard typing, microphone and web camera noise are described and analysed. The analysis focuses on randomness, usability and volume of gathered data. Also the Fortuna random number generator is described from the theoretical view. Object analysis and implementation details are described in the last chapter of the document.
|
|
|
Authentication using a mobile phone
Križan, Viliam ; Hajný, Jan (referee) ; Malina, Lukáš (advisor)
Bachelor’s thesis describes the basics of authentication methods by knowledge, used authentication tokens and biometric ways of authentication. It discusses the capabilities of the individual methods designed for mobile phones. It covers the basic security in the GSM network (using SIM cards) and usage of authentication based on passwords, PIN codes, gestures, facial recognition and speech recognition. Thesis evaluates individual methods in terms of success and performance. It discusses the usage possibilities of authentication methods on the Android platform, where it provides the options of modular arithmetic operations of cryptographic primitives. The work evaluates the time consumption of modular arithmetic operations on the mobile phones with created application. It also lists available algorithms of symmetric and asymmetric ciphers, one-way hashes on the Android platform. The section of authentication protocol conception writes about protocol based on zero-knowledge proof, which authenticates users anonymously. It also proposes the transfer of calculated authentication variables using NFC technology. The protocol also supports the revocation of malicious users. The goal of the thesis was to create two applications that have implemented the selected authentication protocol. The first application designed for users, runs on mobile phones with Android operating system. Application calculates the authentication variables using authentication parameter which it loads from secure element of the phone. After calculations it sends the authentication variables via NFC reader (connection phone – reader). NFC reader is connected to the second application, which verifies the user’s authentication variables and decides the conclusion of authentication. The last part of thesis is devoted to testing the developed applications and running them on the devices. It evaluates the security of the authentication protocol, usage of the phone’s secure element and a security of the transmission over NFC.
|
|
|
Processes Supporting Decision-Making
Križan, Viliam ; Uher, Václav (referee) ; Karásek, Jan (advisor)
This thesis deals with algorithms for supporting decision processes. Firstly, Analytic Hierachy Process (AHP) and Analytic Network Process (ANP) developed by prof. Thomas L. Saaty are described. Basic principles and the implementation in Java programing language of the decision processes are explained. Both processes are then analyzed and their pros and cons and also practical aspect are explained. Secondly, the genetic algorithm NSGA-II developed by Kalyanmoy Deb is described. There are also basic principles and implementation in Java programing language explained. There are finaly the results of NSAG-II algorithm presented.
|
|
|
Random number generator
Križan, Viliam ; Burda, Karel (referee) ; Ležák, Petr (advisor)
This master thesis deals with a generation of random numbers and Fortuna generator implementation in Java language. In the first part the theoretical familiarization to the issues is introduced. Various entropy sources like mouse movement, keyboard typing, microphone and web camera noise are described and analysed. The analysis focuses on randomness, usability and volume of gathered data. Also the Fortuna random number generator is described from the theoretical view. Object analysis and implementation details are described in the last chapter of the document.
|
|
|
Social Network Analysis using methods of pattern recognition
Križan, Viliam ; Burget, Radim (referee) ; Atassi, Hicham (advisor)
Diplomová práca sa zaoberá rozpoznávaním emócií z textu v sociálnych sieťach. Práca popisuje súčasné metódy extrakcie príznakov, používané lexikóny, korpusy a klasifikátory. Emócie boli rozpoznávané na základe klasifikátoru, netrénovaného na anotovaných dátach z mikroblogovacej siete Twitter. Výhodou použitia služby Twitter, bolo geografické vymedzenie dát, ktoré umožňuje sledovanie zmien emócií populácie v rôznych mestách. Prvým prístupom klasifikácie bolo vytvorenie Baseline algoritmu, ktorý používal jednoduchý lexikón. Pre zlepšenie klasifikácie sme v druhom bode použili komplexnejší SVM klasifikátor. SVM klasifikátory, extrakcie a selekcie príznakov boli použité z dostupnej Python knižnice Scikit. Dáta pre natrénovanie klasifikátoru boli zhromažďované z oblasti USA, a to s pomocou vytvorenej aplikácie. Klasifikátor bol natrénovaný na dátach, označených pri ich zhromažďovaní - bez manuálnej anotácie. Boli použité dve rôzne implantácie SVM klasifikátorov. Výsledné klasifikované emócie, v rôznych mestách a dňoch, boli zobrazené v podobe farebných značiek na mape.
|
guest ::
