Information security (InfoSec) policies are widely used by institutions as a form of InfoSec control measure to protect their information assets. InfoSec policies are commonly documented in natural language, which is prone to ambiguity and misinterpretation, thereby making it hard, if not impossible, for users to comply with. These misinterpretations may lead the students or staff members to wrongfully execute the required actions, thereby making institutions vulnerable to InfoSec attacks. According to the literature review conducted in this work, InfoSec policy documents are often not followed or complied with; and the key issues facing InfoSec policy compliance include the lack of management support for InfoSec, organisational cultures of non-compliance, intentional and unintentional policy violation by employees (the insider threat), lack of policy awareness and training as well as the policy being unclear or ambiguous. This study is set in the higher education context and explores the extent to which the non-compliance problem is embedded within the policy documents themselves being affected by ambiguity. A qualitative method with a case study research strategy was followed in the research, in the form of an inductive approach with a cross-sectional time horizon, whereby a selection case of relevant institutional InfoSec policies were analysed. The data was collected in the form of academic literature and InfoSec policies of higher education institutions to derive themes for data analysis. A qualitative content analysis was performed on the policies, which identified ambiguity problems in the data. The findings indicated the presence of ambiguity within the policy documents, making it possible to misinterpret some of the policy statements. Formal methods were explored as a possible solution to the policy ambiguity. A framework was then proposed to address ambiguity and improve on the clarity of the semantics of policy statements. The framework can be used by policy writers in paying attention to the presence of ambiguity in their policies and address these when drafting or revising their policy documents. / School of Computing

The proliferation of the internet and associated online activities exposes users to numerous information security (InfoSec) threats. Such online activities attract a variety of online users who include novice computer users with no basic InfoSec awareness knowledge. Information systems that collect and use sensitive and confidential personal information of users need to provide reliable protection mechanisms to safeguard this information. Given the constant user involvement in these systems and the notion of users being the weakest link in the InfoSec chain, technical solutions alone are insufficient. The usability of online InfoSec systems can play an integral role in making sure that users use the applications effectively, thereby improving the overall security of the applications. The development of online InfoSec systems calls for addressing the InfoSec problem as a social problem, and such development must seek to find a balance between technical and social aspects. The research addressed the problem of usable security in online InfoSec applications by using an approach that enabled the consideration of both InfoSec and usability in viewing the system as a socio-technical system with technical and social sub-systems. Therefore, the research proposed a socio-technical framework that promotes the development of usable security for online information systems using online banking as a case study. Using a convergent mixed methods research (MMR) design, the research collected data from online banking users through a survey and obtained the views of online banking developers through unstructured interviews. The findings from the two research methods contributed to the selection of 12 usable security design principles proposed in the sociotechnical information security (STInfoSec) framework. The research contributed to online InfoSec systems theory by developing a validated STInfoSec framework that went through an evaluation process by seven field experts. Although intended for online banking, the framework can be applied to other similar online InfoSec applications, with minimum adaptation. The STInfoSec framework provides checklist items that allow for easy application during the development process. The checklist items can also be used to evaluate existing online banking websites to identify possible usable security problems. / Computer Science / D. Phil. (Computer Science)

Phishing (nätfiske) fortsätter att vara ett av de vanligaste hoten för användare på Internet. På grund av detta så har mycket forskning gjorts på säkerhetsåtgärder för att identifiera och stoppa nätfiske. Mycket av detta arbete går till maskininlärning, medans ett välkänt behov av utbildning av användarsäkerhet på Internet finns. Användare är den största sårbarheten inom IT, och de borde därför bli utbildade och uppmuntrade att agera säkert på Internet. Denna studie fokuserade på ett mellanstort företag som jobbar med IT i Sverige, på dess användares förmåga och vilja att hantera sin epost säkert för att undvika att gå på nätfiske. Detta har studerats genom intervjuer med anställda på företaget. För möjlighet för transparens och replikering av studiens resultat så startades intervjuerna med att etablera respondenternas bakgrund och erfarenhet med nätfiske. Detta följdes av en demonstration av de vanligaste tecknen på nätfiske som hölls för varje intervjuade individ, för att försäkra att de hade kunskap om dem. Demonstrationen följdes sedan av ytterligare frågor som uppmuntrade deltagarna att först reflektera på ämnet, och sedan möjligheter för dem att utöka sin kunskap.Det som kom fram under intervjuerna var att användarna på företaget alla hade erfarenhet av nätfiske och visste i teorin hur man identifierar nätfiske. Detta trots att majoriteten av medverkande inte hade någon officiell utbildning inom ämnet. Alla användare som medverkade var villiga att göra det som förväntas av dem ifrån företaget. Alla respondenter i denna studie var villiga att göra det de kan för att undvika nätfiske, och några var villiga att gå längre än så på deras arbetsplats genom att utöka deras kompetens. Ingen medverkande i studien uttryckte ovilja att utbilda sig själv ytterligare i ämnet. Två av de åtta respondenterna tog emellertid upp ett starkt argument, att tiden för kontrollen av eposten inte borde överskrida användbarheten av processen. Sammanfattningsvis så har vissa användare behov av incentiv för att förbättra sin säkerhet, någon form av förklaring för varför vissa åtgärder behövs. Om sådant rättfärdigande kunde göras så fanns inga klagomål eller motvilja till att agera mer säkert. / Phishing continues to be one of the most common threats for users of the Internet. As such, a lot of research is made into security measures to identify and stop phishing. A lot of this work goes into machine learning, while it is known that user education on Internet security is needed. Users are the biggest vulnerability within IT, and should therefore be educated and encouraged to act securely on the Internet. This study's focus is on a medium-sized company working with IT in Sweden, on their users’ ability and willingness to handle their emails securely to avoid falling for phishing scams.This was studied through interviews with employees of the company. For the sake of transparency and replication, the interviews were started by establishing the respondents’ background and experiences with phishing. Following this a demonstration of some of the most common tells of phishing was held for each interviewee, to ensure that they know about them. The demonstration was then followed by further questions encouraging the interviewees to reflect upon first the subject, then possible opportunities for them to further their knowledge.What was found is that the users of the company all had experience with phishing and were knowledgeable in theory about how to identify phishing. This is despite the majority of the participants having no official education on the subject. It was found that all users who participated were willing to do what they are expected to do from the company. All respondents in this study were willing to do what they could to avoid phishing, and some were willing to go beyond that at their workplace by expanding their skills. No participant in the study expressed unwillingness to educate themselves further on the subject. However, an important opinion raised by two of the eight respondents was the aspect of time consumed to control all emails should not overshadow the usefulness of the practice. It was concluded that some users needed incentives to improve their security, justifications of why certain measures were needed. If such justifications could be made, there were no complaints or reluctance to act more securely.