This thesis investigates the interplay between international politics and cyberspace, and explains how Taiwan's cybersecurity policy was formed prior to 2016. It examines how politics can shape or reshape the future direction of technologies. By using Taiwan as the object of the case study, the central research question is, "How is Taiwan counterbalancing China's rising power in cyberspace and what are the implications?" The investigation for the first part of the research question provides a general account of issues affecting Taiwan's practice of cybersecurity policy via a constructivist approach. While I do not deny the technology determinist's logic that new technology can drive the way of politics, the empirical observations from Taiwan focus our attention on a different perspective, that politics can still reshape future direction and the use of technology. It explains to us through the case of Taiwan how politics trump both technical decisions and direction of technology, and further exposes the underlying knowledge structure within Taiwanese policy makers' "world." While this knowledge structure, as a form of theory, constitutes the world we live in, the second part of the research question scrutinizes this taken-for-granted knowledge structure. I challenge well-accepted assumptions regarding cyberwarfare to investigate its limitations and explore its problematic effects within the context of Taiwan. By using the principles of Critical Security Studies, I argue that an alternative conceptualization of cybersecurity can still embrace the security end that Taiwan intends to achieve, and propose a critical strategy to engage Taiwan's security challenge while avoiding adverse consequences from cyberwarfare. Looking closely at the case of Taiwan's cybersecurity contributes to the broader International Relations (IR) literature concerning the effects of norms and interest, and extends a constructivist approach to the domain of cyberspace. It also allows knowledge in Cybersecurity Studies to establish a dialogue with IR literature, and reduces the knowledge gap of Taiwan's cybersecurity in Taiwan studies. This project was conducted via interdisciplinary approaches situated at the intersection of IR, Cybersecurity Studies, and Taiwan Studies, and is a timely reminder of the need to examine Taiwan's security with a more contemporary, localized, and theoretically-grounded framework that will help policymakers understand the new challenges that they face in the 21st Century. It is a discourse of resistance to the current discussions that centre on cyberwarfare and instead turns our attention to true cybersecurity.

This diploma thesis deals with cyber relation of the United States of America and the People's republic of China. The aim of this diploma thesis is to answer the following questions: What kind of cyber threat for the United States does China represent? How is China's cyber strategy characterised? How do USA react on this threat and what are the gaps in this reaction? The thesis is divided into four chapters. The first chapter deals with definition of cyberwarfare and its perception in Chinese context. The second chapter analyses USA - China relation and its implication for cyber security. The third chapter represents US reaction on Chinese cyber threat. The last chapter deals with the gaps in the reaction. Keywords USA, China, cyber threat, cyberwarfare, cyber espionage

I en alltmer digitaliserad värld blir den virtuella verkligheten lika viktig som den fysiska. Sverige ligger i framkant när det kommer till digitalisering, men samtidigt halkar landet efter i cybersäkerhet. Sammantaget kan det innebära att krig även kan förekomma i den virtuella verkligheten. I svensk media har begreppet cyberkrig fått en ökad uppmärksamhet men det har visat sig att begreppet inte har en gemensam definition i Sverige. Vilka anledningar ligger bakom det, hur kan cyberkrig alternativt definieras och hur förhåller sig begreppet cyberkrig till statliga myndigheter ur ett informations-och cybersäkerhetsperspektiv i en alltmer digitaliserad värld?  Denna kvalitativa studie syftar till att utreda dessa frågetecken genom att delge privata och statliga aktörers perspektiv för att bidra med en fördjupad förståelse för begreppet cyberkrig. Studiens litteratur- och artikelstudie, samt semistrukturerade intervjuer ger en möjlighet att undersöka individuella perspektiv, doktriner och rapporter från olika myndigheter i både Sverige och internationellt. Utifrån resultatet i detta arbete dras slutsatsen att begreppet cyberkrig används i det offentliga rummet men att det inte nödvändigtvis brukas bland statliga aktörer och i formella sammanhang. Statliga aktörer såsom Försvarsmakten, MSB och FOI föredrar i stället att använda sig av begreppet cyberoperationer.Ett annat begrepp som visat sig vara fundamental för att kunna skapa en förståelse för begreppet cyberkrig är cyberdomänen. Författarna till detta arbete har tillsammans framställt en alternativ definition av cyberkrig enligt följande ”Cyberkrig innebär att en nation eller organisation inom cyberdomänen angriper eller försöker skada en annan nations digitala infrastruktur. Konsekvenserna ska kunna likställas med ett väpnat angrepp i enlighet med artikel 2(4) i FN-stadgan”. / In an increasingly digitalized world, the virtual reality is becoming as significant as the physical one. Sweden is at the forefront of digitalization, but at the same time, it is slipping behind in cybersecurity. Together, this may mean that warfare also can occur in virtual reality. The concept of cyberwarfare has received increased attention in Swedish media; however, it has become clear that the concept does not have a definition in Sweden. What are the reasons behind this, how can cyberwarfare be alternatively defined, and how does the concept of cyberwarfare relate to Swedish authorities’ defense work in an increasingly digitalized world? This qualitative study aims to investigate these questions by sharing the perspectives of private and state actors to contribute to a deeper understanding of the concept of cyberwarfare. This study’s literature and article study, as well as semi-structured interviews, provide an opportunity to examine individuals’ perspectives and other material such as doctrines and reports from various authorities in Sweden and internationally. Based on the results of this work, it is concluded that the concept of cyberwarfare is used in the public sphere, but not in a formal context and among state actors. State actors such as the Swedish Armed Forces, MSB, and FOI prefer to use the concept of cyberoperations. Another concept that has proved to be fundamental to creating an understanding of the concept of cyber war is the cyber domain. The authors of this study have defined cyberwarfare as when “a nation or organization in the cyber domain attacks or attempts to damage the digital infrastructure of another nation. The consequences must be equivalent to an armed attack following Article 2(4) of the UN Charter”.

This thesis explores to what extent Multinomial Naive Bayes (MNB) and Support Vector Machines (SVM) classifiers can be used to determine the polarity of news, specifically the news coverage of Sweden by the Russian state-funded news outlets RT and Sputnik. Three experiments are conducted.  In the first experiment, an MNB and an SVM classifier are trained with the Large Movie Review Dataset (Maas et al., 2011) with a varying number of samples to determine how training data size affects classifier performance.  In the second experiment, the classifiers are trained with 300 positive, negative, and neutral news articles (Agarwal et al., 2019) and tested on 95 RT and Sputnik news articles about Sweden (Bengtsson, 2019) to determine if the domain specificity of the training data outweighs its limited size.  In the third experiment, the movie-trained classifiers are put up against the domain-specific classifiers to determine if well-trained classifiers from another domain perform better than relatively untrained, domain-specific classifiers.  Four different types of feature sets (unigrams, unigrams without stop words removal, bigrams, trigrams) were used in the experiments. Some of the model parameters (TF-IDF vs. feature count and SVM’s C parameter) were optimized with 10-fold cross-validation.  Other than the superior performance of SVM, the results highlight the need for comprehensive and domain-specific training data when conducting machine learning tasks, as well as the benefits of feature engineering, and to a limited extent, the removal of stop words. Interestingly, the classifiers performed the best on the negative news articles, which made up most of the test set (and possibly of Russian news coverage of Sweden in general).