The proposed work focuses on articulating a mathematical framework to capture the structure of an ontology and relate it to organised data sets. In the discussed framework, the ontology structure captures the mereological relationships between concepts. It also uses other relationships relevant to the considered domain of application. The organised dataset component of the framework is represented using diagonal-free cylindric algebra. The proposed framework, called the domain-information structure, enables us to link concepts to data sets through a number of typed data operators. The new framework enhances concurrent reasoning on data for knowledge generation, which is essential for handling big data. We illustrate the advantage of the obtained framework by using it in generating new knowledge from an ontology and a given data set.

Thesis

Master of Applied Science (MASc)

With the massive deluges of data that several domains of study experience, referred to as Big Data, the need to efficiently process and analyze data has risen. Ontologies have been employed to structure these domains with the ultimate purpose of generating new knowledge about the respective domain. Currently, when creating or examining an ontology, the concepts of the domain are limited to being statically defined in relation to other concepts. This limitation on the concept's definition affects the reasoning process by omitting or not properly representing all information that may exist in the domain such as how the constantly evolving environment changes how the data can be understood. The resulting tool of this research, Conto, allows for the interpretation of a concept as an abstract data type. When coupled to knowledge generation process, these interpretations allow the obtention of new knowledge that would traditionally be unobtainable. The different types of knowledge that can be obtained via the multiple interpretations are explored in this work using examples of ontologies. Conto is a Protege plugin that uses Ontograf to display the ontologies.

Thesis

Master of Applied Science (MASc)

The set of related products is referred to as a product family, and feature-modeling is a widely used technique to capture the commonalities and variabilities of a product family in terms of "features". With the growing complexity of software product families in several software industries, the development, maintenance and evolution of complex and large feature models are among the main challenges faced by feature-modeling practitioners. In particular, more sophisticated feature-modeling techniques are required to address the problems caused by unanticipated changes and crosscutting concerns in feature models. This thesis tackles the above challenges in feature-modeling by adopting the aspect-oriented paradigm at the feature-modeling level. I first introduce a specification language, called AO-PFA, which is an extension of the Product Family Algebra (PFA) language. I then proposed a formal verification technique to check the compatibility of aspects with their base specifications in AO-PFA. In the aspect-oriented paradigm, the process of combining aspects with base specifications is referred to as the weaving process. I finally discussed how to perform the weaving process in AO-PFA. By proposing a systematic approach to extend product family algebra with the abilities of specifying, verifying, and weaving aspects, we are able to handle the difficulties that arise from crosscutting concerns and unanticipated changes in large-scale feature models.

Doctor of Philosophy (PhD)

Feature modelling is an emerging software engineering paradigm, which helps organizations to develop products from core assets. Products are organized into families that share common core features. Feature modelling involves capturing, into a feature model, the commonality and variability of product families and several relationships among features or products. This thesis is about proposing a language for specifying feature models that is based on product family algebra (PFA). The language is intended to encompass the constructs found in early feature modelling graphical notations and languages. The thesis gives the syntax and the semantics of the proposed language. It discusses the design of its compiler that takes a feature model specification and generates its corresponding PFA, which can be analyzed using the tool Jory. The thesis uses a quite extensive case study to illustrate the use of the proposed language and its compiler.

Master of Applied Science (MASc)

Covert channels are means of communication that allow agents in a system to transfer information in a manner that violates the system’s security policy. Covert channels have been well studied in the constrained and old sense of the term where two agents are communicating through a channel while an intruder interferes to hide the transmission of a message. In an increasingly connected world where modern computer systems consist of broad and heterogeneous communication networks with many interacting agents, distributed covert channels are becoming increasingly available. For these distributed forms of covert channels, there are shortcomings in the science, mathematics, fundamental theory, and tools for risk assessment, and for proposing mechanisms and design solutions for averting these threats. Since current formal methods for specifying concurrent systems do not provide the tools needed to efficiently tackle the problem of distributed covert channels in systems of communicating agents, this thesis proposes Communicating Concurrent Kleene Algebra (C²KA) which is an extension to the algebraic model of concurrent Kleene algebra (CKA) first presented by Hoare et al. C²KA is used to capture and study the behaviour of agents, and description logic is used to capture and study the knowledge of agents. Using this representation of agents in systems of communicating agents, this thesis presents a formulation and verification approach for the necessary conditions for the existence of distributed covert channels in systems of communicating agents. In this way, this thesis establishes a mathematical framework for the modelling, analysis, and mitigation of distributed covert channels in systems of communicating agents. This framework enhances the understanding of covert channels and provides a basis for thinking and reasoning about covert channels in new ways. This can lead to a formal foundation upon which guidelines and mechanisms for designing and implementing systems of communicating agents that are resilient to covert channels can be devised.

Thesis

Doctor of Philosophy (PhD)

Modelling nonfunctional requirements, along with many other NFR-related concerns, have not been addressed properly in the literature. Although nonfunctional requirements (or quality attributes) are important, they are the most expensive and difficult to deal with since they are (mostly) specified qualitatively not quantitatively, and also due to the fact that nonfunctional requirements may have interdependencies among each other leading to inconsistency in requirements specification. Moreover, the adoption of the concept of product families into the software industry led to the ability today to build families that share features. This indicates the significance of software families. Accordingly, in this research, not only do we study nonfunctional requirements in a systematic way, we also attempt to examine them from the perspective of software families. We highlight the need for developing a better quantitative assessment technique for nonfunctional requirements. Then, we propose a formal approach to the assessment of nonfunctional requirements in software product families.

Thesis

Master of Applied Science (MASc)

With the emergence of computers in every day activities and with the ever-growing complexity of networks and network protocols, covert channels are becoming an eminent threat to confidentiality of information. With increasing sensitivity of data in many computer application domains, the leakage of confidential information can have severe repercussions on the institution from which the information was leaked. In light of this eminent threat, we propose a technique to detect confidential information leakage via covert channels. We limit our focus to instances where the users of covert channels modulate the information that is being sent; either by encryption, or some other form of encoding. In the literature, the difference between classes of covert channels under the current classification is unclear. This lack of clarity results in the development of incomplete techniques for modeling, detecting and preventing covert channels. In this thesis, we propose a new classification for covert channels which organizes covert channels into two types: protocol-based covert channels and environment-based covert channels. We also develop a novel, comprehensive model for protocol-based covert channel communication. Using the developed model, we explore the relationship between covert channel communication, steganography and watermarking. The intent is to provide a better understanding of covert channel communication in an attempt to develop investigative support for confidentiality. Finally, we propose a technique for detecting confidential information leakages via covert channels. The technique is based on relation algebra and offers tests for verifying the existence of an abstraction relation which relates the confidential information to the information that is observed to be sent on the communication channel. It focuses on protocol-based covert channels. With a better understanding of covert channel communication, we are able to develop more effective and efficient mechanisms for detecting and preventing the use of covert channels to leak confidential information in computer systems.

Master of Applied Science (MASc)

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.5px Times} Feature models are commonly used to capture the commonalities and the variability of product families. There are several feature modelling notations that correspondingly depict the concepts of feature modelling techniques. Therefore, the tools based on them reflect this diversity in the notations used and the fuzziness of the concepts adopted. The thesis discusses the design and the construction of a tool that is based on Product-Families Algebra (PFA) and on Binary Decision Diagrams (BDD). The first brings the mathematical formalism to the specifications of product families and the mathematical theory that enables calculations on feature-models. The second brings efficient algorithms in time and in space. Hence, the tool allows several algebraic manipulations of feature models that are specified within the language of PFA. We coined this tool Jory. The main contribution of the thesis is the design of the tool, and the implementation of four layers of its architectural design. As well, the thesis gives an implementation of multi-sets and the operations on them using BDDs. Several case studies are presented and used in the validation of the tool. p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.5px Times}

Master of Science (MS)