Thesis (Ph.D.)--State University of New York at Albany, 2007. / Includes bibliographical references.

Thesis (M.A. in Security Studies (Homeland Security and Defense))--Naval Postgraduate School, September 2009. / Thesis Advisor(s): Bellavita, Christopher. "September 2009." Description based on title screen as viewed on 5 November 2009. Author(s) subject terms: Homeland security, collaborative capacity, federalism, grants, homeland security stakeholders, Australian Security Intelligence Organization (AISO), domestic intelligence agency, information sharing, Virginia State Police. Includes bibliographical references (p. 101-110). Also available in print.

Thesis (M.S. in Computer Science)--Naval Postgraduate School, Sept. 2004. / Thesis advisor(s): Cynthia E. Irvine, Karen Burke. Includes bibliographical references (p. 161-162). Also available online.

Thesis (M. S.)--Florida State University, 2004. / Advisor: Dr. Mike Burmester, Florida State University, College of Arts and Sciences, Dept. of Computer Science. Title and description from dissertation home page (viewed Jan. 18, 2005). Includes bibliographical references.

Thesis (Ph. D.)--Rutgers University, 2008. / "Graduate Program in Political Science." Includes bibliographical references (p. 226-240).

Thesis (Ph. D.)--Purdue University, 1994. / Vita. Includes bibliographical references (leaves 289-302).

Thesis (M.S. in Computer Science)--Naval Postgraduate School, June 2005. / Thesis Advisor(s): Neil Rowe, J.D. Fulp. Includes bibliographical references (p. 53-54). Also available online.

Thesis (M.S.)--Brigham Young University. Dept. of Computer Science, 2006. / Includes bibliographical references (p. 41-44).

Users share a large amount of information with modern platforms such as web platforms and social platforms for various services. However, they face the risk of information leakage because modern platforms still lack proper security policies. Existing security policies, such as permission systems and isolation, can help regulate information sharing. However, these policies have problems, such as coarse granularity, bad usability, and incompleteness, especially when new features are introduced. I investigate the security impacts of new features in web and mobile platforms and find design problems that lead to user information leakage. Based on these analyses, I propose design principles for permission systems that mediate how information should be shared in modern and emerging platforms, such as web and social platforms, to provide functionality with privacy preserved. I aim to design permission systems that only allow least-privilege information access. Specifically, I utilize program analysis and natural language processing to understand how applications use sensitive data and correlate these data with their functionality. With this understanding, I design schemes that ask for user consent about unexpected information access and automatically reduce overprivileged access. I provide guidelines for platform designers to build their permission systems according to respective adversary models and resources. In particular, I implement the new permission system for social platforms and Internet of Things (IoT) platforms that enable least-privilege information sharing. For the social platforms, I incorporate the primitives of Opaque handle, Opaque display, and User-driven access control (OOU) to design a least-privilege, user-friendly, developer-friendly, and feature-rich permission system. According to my study on Facebook, OOU can be applied to remove or replace 81.2% of sensitive permission instances without affecting functionality. For IoT platforms, I present a new authorization framework, SmartAuth, that supports user-centric, semantic-based authorization. SmartAuth automatically collects security-relevant information from an IoT application’s description, code, and annotations, and generates an authorization user interface to bridge the gap between the functionalities explained to the user and the operations the application actually performs.

E-Government security is a major area of concern that has the potential to affect the success of e-Government services across the world. Much of the literature has addressed this phenomenon by applying principles of computer science or engineering which tend to be objective. User concern of e-Government service security has not been addressed applying social science principles or management that tend to be subjective and have not been addressed in the literature. Objective research outcomes are unfortunately not suitable to address subjective factors. Further, user centric approach has not been adopted in most of the empirical studies that have dealt with e-Government security leading to lack of an understanding of how users perceive or feel or comprehend about e-Government services, particularly e-Government service security. Most of the research efforts addressing e-Government security have focused on either technological issues or engineering issues neglecting user perceptions and behavioural aspects. This disadvantage has led to possible reduction in the up-take of e-Government services. There was a need to have an in-depth understanding of user centric e-Government security and user centric factors that affect it as its antecedents addressing which it is possible to enhance user confidence in e-Government and hence its success. This research has addressed this partially. While addressing the concerns raised above, this research has defined and identified certain user centric factors that are required to examine the user centric nature of e-Government service security from the management and social sciences perspective. E-Government literature was critically reviewed to determine the user centric factors and their relationship to user centric e-Government security with the help of theories, models, concepts and frameworks that have not been applied so far. Contextual factors have been identified as important user centric ones that affect user centric e-Government security with e-Government technology chosen as the main contextual determinant of user centric e-Government security. User trust and user felt risk in using e-Government services were brought in as mediators of this relationship due to the prime importance these two user centric factors carry with regard to affecting the relationship between technology and user centric e-Government security. In addition demographic factors and culture (nationality) as a factor were applied to test their influence on the relationship between user trust and user centric e-Government security mediated by user felt risk to find whether they have any impact. Moderators (Human Computer Interaction (HCI), user privacy and web design quality) of this relationship were added to the investigation as literature showed that e-Government technology could not operate in isolation. Finally empirical outcomes of testing the above relationships were practically tested by examining the influence of perceived ease of use and usefulness on the relationship between user trust and user centric e-Government security mediated by user felt risk to find whether technology impacted users in reality. Theoretical framework was drawn from the literature review leading to a conceptual model that was used to answer the research question. 12 hypotheses were tested in all. The research was conducted in the Kingdom of Bahrain which ranks high in the implementation of e-Government (e.g. 14th ranked in the world in implementing e-participation in 2014 ranked by UN). The country offered a fertile ground for conducting research as the e-Government service provided were updated technologically constantly with the latest technological advancement cloud computing introduced in e-Government service provision. Most government services were offered now through e-Government services. The population was cosmopolitan and education levels of the users of e-Government were reasonably high providing a strong basis for conducting this research. Quantitative research method and survey questionnaire strategy were used. Users of e-Government services were the target population. Sampling procedure yielded 309 valid responses. Rigourous statistical analysis provided the findings. Except for 2 hypotheses the remaining were verified and established. Technology was found to determine user centric e-Government security with the mediation by trust being stronger than risk. HCI and web design quality moderated the relationship between technology and user centric e-Government security significantly. User education and experience were found to influence user trust and user centric e-Government security. User privacy and nationality were not found to be statistically significant. Perceived ease of use and usefulness of the technology were found to influence e-Government security mediated by trust and risk. This research was perhaps one of the first to have been conducted in a context where e-Government technology used cloud computing. The research contributed to the growing body of knowledge in the field of e-Government security that has viewed this phenomenon from the lens of social sciences and management. Theoretical contribution showed how the operationalization and relationship amongst the factors could be explained by expanding the application of theories including socio-technical, behavioural, managerial, technology adoption, organiational and HCI. Practical implications showed the usefulness of this research to users, service providers and policy makers involved with e-Government services. Methodologically this research has introduced a verification stage by which it has verified the theoretical results using practical outcomes.