CEC Theses and Dissertations

Date of Award

2014

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Information Systems (DISS)

Department

Graduate School of Computer and Information Sciences

Advisor

James D Cannady

Committee Member

Rayford Vaughn

Committee Member

Joseph Gulla

Abstract

Information Systems today rarely are contained within a single user workstation, server, or networked environment. Data can be transparently accessed from any location, and maintained across various network infrastructures. Cloud computing paradigms commoditize the hardware and software environments and allow an enterprise to lease computing resources by the hour, minute, or number of instances required to complete a processing task. An access control policy mediates access requests between authorized users of an information system and the system's resources. Access control policies are defined at any given level of abstraction, such as the file, directory, system, or network, and can be instantiated in layers of increasing (or decreasing) abstraction. For the system end-user, the functional allocation of security policy to discrete system components, or subsystems, may be too complex for comprehension. In this dissertation, the concept of a metapolicy, or policy that governs execution of subordinate security policies, is introduced. From the user's perspective, the metapolicy provides the rules for system governance that are functionally applied across the system's components for policy enforcement. The metapolicy provides a method to communicate updated higher-level policy information to all components of a system; it minimizes the overhead associated with access control decisions by making access decisions at the highest level possible in the policy hierarchy. Formal definitions of policy often involve mathematical proof, formal logic, or set theoretic notation. Such policy definitions may be beyond the capability of a system user who simply wants to control information sharing. For thousands of years, mankind has used narrative and storytelling as a way to convey knowledge. This dissertation discusses how the concepts of storytelling can be embodied in computational narrative and used as a top-level requirements specification. The definition of metapolicy is further discussed, as is the relationship between the metapolicy and various access control mechanisms. The use of storytelling to derive the metapolicy and its applicability to formal requirements definition is discussed. The author's hypothesis on the use of narrative to explain security policy to the system user is validated through the use of a series of survey instruments. The survey instrument applies either a traditional requirements specification language or a brief narrative to describe a security policy and asks the subject to interpret the statements. The results of this research are promising and reflect a synthesis of the disciplines of neuroscience, security, and formal methods to present a potentially more comprehensible knowledge representation of security policy.

Share

COinS