As machine learning becomes an increasingly relevant field being incorporated into everyday life, so does the need for consistently high performing models. With these high expectations, along with potentially restrictive data sets, it is crucial to be able to use techniques for machine learning that increase the likelihood of success. Robust Principal Component Analysis (RPCA) not only extracts anomalous data, but also finds correlations among the given features in a data set, in which these correlations can themselves be used as features. By taking a novel approach to utilizing the output from RPCA, we address how our method effects the performance of such models. We take into account the efficiency of our approach, and use projectors to enable our method to have a 99.79% faster run time. We apply our method primarily to cyber security data sets, though we also investigate the effects on data sets from other fields (e.g. medical).

abstract: Due to the increase in computer and database dependency, the damage caused by malicious codes increases. Moreover, gravity and the magnitude of malicious attacks by hackers grow at an unprecedented rate. A key challenge lies on detecting such malicious attacks and codes in real-time by the use of existing methods, such as a signature-based detection approach. To this end, computer scientists have attempted to classify heterogeneous types of malware on the basis of their observable characteristics. Existing literature focuses on classifying binary codes, due to the greater accessibility of malware binary than source code. Also, for the improved speed and scalability, machine learning-based approaches are widely used. Despite such merits, the machine learning-based approach critically lacks the interpretability of its outcome, thus restricts understandings of why a given code belongs to a particular type of malicious malware and, importantly, why some portions of a code are reused very often by hackers. In this light, this study aims to enhance understanding of malware by directly investigating reused codes and uncovering their characteristics. To examine reused codes in malware, both malware with source code and malware with binary code are considered in this thesis. For malware with source code, reused code chunks in the Mirai botnet. This study lists frequently reused code chunks and analyzes the characteristics and location of the code. For malware with binary code, this study performs reverse engineering on the binary code for human readers to comprehend, visually inspects reused codes in binary ransomware code, and illustrates the functionality of the reused codes on the basis of similar behaviors and tactics. This study makes a novel contribution to the literature by directly investigating the characteristics of reused code in malware. The findings of the study can help cybersecurity practitioners and scholars increase the performance of malware classification. / Dissertation/Thesis / Masters Thesis Computer Science 2020

The master’s thesis is focused on the design and creation of a solution for converting company’s software application into the mobile and web form. The main goal is make business processes more efficient and maintain information and cyber security. The result should be a system that brings an innovative and convenient solution, time and financial savings.

Deepfake technológia je v poslednej dobe na vzostupe. Vzniká mnoho techník a nástrojov pre tvorbu deepfake médií a začínajú sa používať ako pre nezákonné tak aj pre prospešné činnosti. Nezákonné použitie vedie k výskumu techník pre detekciu deepfake médií a ich neustálemu zlepšovaniu, takisto ako k potrebe vzdelávať širokú verejnosť o nástrahách, ktoré táto technológia prináša. Jedna z málo preskúmaných oblastí škodlivého použitia je používanie deepfake pre oklamanie systémov hlasovej autentifikácie. Názory spoločnosti na vykonateľnosť takýchto útokov sa líšia, no existuje len málo vedeckých dôkazov. Cieľom tejto práce je preskúmať aktuálnu pripravenosť systémov hlasovej biometrie čeliť deepfake nahrávkam. Vykonané experimenty ukazujú, že systémy hlasovej biometrie sú zraniteľné pomocou deepfake nahrávok. Napriek tomu, že skoro všetky verejne dostupné nástroje a modely sú určené pre syntézu anglického jazyka, v tejto práci ukazujem, že syntéza hlasu v akomkoľvek jazyku nie je veľmi náročná. Nakoniec navrhujem riešenie pre zníženie rizika ktoré deepfake nahrávky predstavujú pre systémy hlasovej biometrie, a to používať overenie hlasu závislé na texte, nakoľko som ukázal, že je odolnejšie proti deepfake nahrávkam.

Cyber security is a major concern due to global incidents of intrusion. The impact of the attacks on the electricity grid can be significant, resulting in the collapsing of the national economy. Electricity network is needed by banks, government security agencies, hospitals and telecommunication operators. The purpose of this research is to investigate the various types of cyber security threats, including ICT technologies required for safe operation of the smart grid to protect and mitigate the impact of cyber security. The modelling of cyber security using the Matlab/SimPowerSystem simulates the City of Tshwane power system. Eskom components used to produce energy, interconnect to the City of Tshwane power distribution substations and simulated using Simulink SimPowerSystem. / Dissertation (MEng)--University of Pretoria, 2017. / Electrical, Electronic and Computer Engineering / MEng / Unrestricted

As the connectivity of home devices elevates so does the volume and sophistication of cyber attacks consistently grow. Therefore, the need for network security and availability becomes more significant. Numerous sorts of countermeasures like firewalls and router-based packet filtering have been put in place, although these alone are not enough to brace the network from unauthorised access. One of the most efficient methods of stopping network adversaries is using Intrusion Detection and Prevention Systems (IDPS). The goal of an IDPS is to stop security attacks before they can be successfully carried out. In this paper, I looked at four network attacks namely; probing, denial of service, remote to user and user to root and improved their respective Snort rules to optimize processing time and capturing capacity using regular expressions and fast pattern. Snort with improved rules captured 100% of the attacks launched to the network while without the improved rules, Snort captured between 0% to 60% of the attacks launched to the network making an improvement of 40%.

While it is evident that network services continue to play an ever-increasing role in our daily lives, it is less evident that our information infrastructure requires a concerted, well-conceived, and fastidiously executed strategy to remain viable. Government agencies, Non-Governmental Organizations (\NGOs"), and private organizations are all targets for malicious online activity. Security has deservedly become a serious focus for organizations that seek to assume a more proactive posture; in order to deal with the many facets of securing their infrastructure. At the same time, the discipline of data science has rapidly grown into a prominent role, as once purely theoretical machine learning algorithms have become practical for implementation. This is especially noteworthy, as principles that now fall neatly into the field of data science has been contemplated for quite some time, and as much as over two hundred years ago. Visionaries like Thomas Bayes [18], Andrey Andreyevich Markov [65], Frank Rosenblatt [88], and so many others made incredible contributions to the field long before the impact of Moore's law [92] would make such theoretical work commonplace for practical use; giving rise to what has come to be known as "Data Science". / Includes bibliography. / Dissertation (Ph.D.)--Florida Atlantic University, 2020. / FAU Electronic Theses and Dissertations Collection

Cyber-attacks against small businesses are on the rise yet small business owners often lack effective strategies to avoid these attacks. The purpose of this qualitative multiple case study was to explore the strategies small business owners use to make cyber-security decisions. Bertalanffy's general systems theory provided the conceptual framework for this study. A purposive sample of 10 small business owners participated in the interview process and shared their decision-making methodologies and influencers. The small business owners were vetted to ensure their strategies were effective through a series of qualification questions. The intent of the research question and corresponding interview questions was to identify strategies that successful small business owners use to make cyber-security decisions. Data analysis consisted of coding keywords, phrases, and sentences from semi structured interviews as well as document analysis. The following themes emerged: government requirements, peer influence, budgetary constraints, commercial standards, and lack of employee involvement. According to the participants, budgetary constraints and peer influence were the most influential factors when making decisions regarding cyber-security strategies. Through exposing small business owners to proven strategies, the implications for social change include a reduction of their small business operating costs and assistance with compliance activities.

Increased internal and external training approaches are elements senior leaders need to know before creating a training plan for security professionals to protect sensitive information. The purpose of this qualitative case study was to explore training strategies telecommunication industry leaders use to ensure security professionals can protect sensitive information. The population consisted of 3 senior leaders in a large telecommunication company located in Dallas, Texas that has a large footprint of securing sensitive information. The conceptual framework on which this study was based was the security risk planning model. Semistructured interviews and document reviews helped to support the findings of this study. Using the thematic approach, 3 major themes emerged. The 3 themes included security training is required for all professionals, different approaches to training are beneficial, and using internal and external training's to complement each other. The findings revealed senior leaders used different variations of training programs to train security professionals on how to protect sensitive information. The senior leaders' highest priority was the ability to ensure all personnel accessing the network received the proper training. The findings may contribute to social change by enhancing area schools' technology programs with evolving cyber security technology, helping kids detect and eradicate threats before any loss of sensitive information occurs.

Internet växer idag konstant och det förväntas finnas fler än 50 miljarder enheter anslutna till internet efter år 2020. Flertalet av dessa enheter kommer vara små, inbäddade enheter som är anslutna och kommunicerar via Internet of Things. Att försäkra att dessa enheter är säkra och skyddade från obehörig åtkomst har varit något som väckt oro ända sedan så kallade botnets visat sig kapabla till att ta över och utnyttja hundratusentals Internet of Things anslutna enheter för att utföra Distributed Denial-of-Service attacker. Målet med denna studie har varit att ställa frågan samt svara på hur stor påverkan Internet of Things enheter har när de utnyttjas för att utföra en Distributed Denial-of-Service attack i ett lokalt trådlöst nätverk. För att besvara denna fråga har denna avhandling forskat kring områden som rör cybersäkerhet, Internet of Things, samt metoder för att utföra Distributed Denial-of-Service attacker. Denna studie har implementerat ett scenario som mäter påverkan vid en Distributed Denial-of-Service attack när upp till sex emulerade Internet of Things enheter som attackerar en ensam offerdator via TCP, UDP och HTTP flood metoder i ett lokalt nätverk. Flertalet test har utförts samt analyserats. Resultatet från denna studie presenteras och jämförs vilket visar att offerdatorn är relativt kapabel till att försvara sig mot TCP och HTTP floods med upp till sex Internet of Things enheter vid respektive attack.  Det implementerade scenariot och metoden är huruvida kapabel till att tungt överbelasta offerdatorn när UDP flood används för samtliga sex Internet of Things enheter. / The internet is constantly growing, we are expecting there to be more than 50 billion devices on the internet past 2020. Many of these devices will be small, embedded devices connected and communicating using the Internet of Things. Keeping these devices secure and protected from unauthorized access has been a raising concern in part due to botnets that have proven capable of exploiting hundreds of thousands of Internet of Things devices to carry out Distributed Denial-of-Service attacks in the past. The objective of this study has been to answer how big of an impact compromised IoT devices might have when exploited to carry out a Distributed Denial-of-Service attack in a Wireless Local Area Network. To answer this question this thesis has done research in the fields concerning cyber-security, the Internet of Things, and methods of distributing Denial-of-Service attacks. This study implements a scenario that measures the impact of a Distributed Denial-of-Service attack utilizing up to six emulated IoT devices that attack a single victim computer using a TCP, UDP or HTTP flood. Several tests have been performed and analyzed. The results from this work are presented and compared and shows that the victim computer is relatively capable of mitigating and defending against the TCP and HTTP flood with up to six utilized IoT devices in each attack. In the implemented scenario and method are however capable of heavily congesting and overwhelming a single victim computer when utilizing a UDP flood with all six IoT devices simultaneously attacking.