This thesis applies a hierarchical latent trait model system to a large quantity of data. The motivation for it was lack of viable approaches to analyse High Throughput Screening datasets which maybe include thousands of data points with high dimensions. We believe that a latent variable model (LTM) with a non-linear mapping from the latent space to the data space is a preferred choice for visualising a complex high-dimensional data set. As a type of latent variable model, the latent trait model can deal with either continuous data or discrete data, which makes it particularly useful in this domain. In addition, with the aid of differential geometry, we can imagine that distribution of data from magnification factor and curvature plots. Rather than obtaining the useful information just from a single plot, a hierarchical LTM arranges a set of LTMs and their corresponding plots in a tree structure. We model the whole data set with a LTM at the top level, which is broken down into clusters at deeper levels of the hierarchy. In this manner, the refined visualisation plots can be displayed in deeper levels and sub-clusters may be found. Hierarchy of LTMs is trained using expectation-maximisation (EM) algorithm to maximise its likelihood with respect to the data sample. Training proceeds interactively in a recursive fashion (top-down). The user subjectively identifies interesting regions on the visualisation plot that they would like to model in a greater detail. At each stage of hierarchical LTM construction, the EM algorithm alternates between the E - and M - step. Another problem that can occur when visualising a large data set is that there may be significant overlaps of data clusters. It is very difficult for the user to judge where centres of regions of interest should be put. We address this problem by employing the minimum message length technique, which can help the user to decide the optimal structure of the model.

For analysing financial time series two main opposing viewpoints exist, either capital markets are completely stochastic and therefore prices follow a random walk, or they are deterministic and consequently predictable. For each of these views a great variety of tools exist with which it can be tried to confirm the hypotheses. Unfortunately, these methods are not well suited for dealing with data characterised in part by both paradigms. This thesis investigates these two approaches in order to model the behaviour of financial time series. In the deterministic framework methods are used to characterise the dimensionality of embedded financial data. The stochastic approach includes here an estimation of the unconditioned and conditional return distributions using parametric, non- and semi-parametric density estimation techniques. Finally, it will be shown how elements from these two approaches could be combined to achieve a more realistic model for financial time series.

This thesis presents an effective methodology for the generation of a simulation which can be used to increase the understanding of viscous fluid processing equipment and aid in their development, design and optimisation. The Hampden RAPRA Torque Rheometer internal batch twin rotor mixer has been simulated with a view to establishing model accuracies, limitations, practicalities and uses. As this research progressed, via the analyses several 'snap-shot' analysis of several rotor configurations using the commercial code Polyflow, it was evident that the model was of some worth and its predictions are in good agreement with the validation experiments, however, several major restrictions were identified. These included poor element form, high man-hour requirements for the construction of each geometry and the absence of the transient term in these models. All, or at least some, of these limitations apply to the numerous attempts to model internal mixes by other researchers and it was clear that there was no generally accepted methodology to provide a practical three-dimensional model which has been adequately validated. This research, unlike others, presents a full complex three-dimensional, transient, non-isothermal, generalised non-Newtonian simulation with wall slip which overcomes these limitations using unmatched ridding and sliding mesh technology adapted from CFX codes. This method yields good element form and, since only one geometry has to be constructed to represent the entire rotor cycle, is extremely beneficial for detailed flow field analysis when used in conjunction with user defined programmes and automatic geometry parameterisation (AGP), and improves accuracy for investigating equipment design and operation conditions. Model validation has been identified as an area which has been neglected by other researchers in this field, especially for time dependent geometries, and has been rigorously pursued in terms of qualitative and quantitative velocity vector analysis of the isothermal, full fill mixing of generalised non-Newtonian fluids, as well as torque comparison, with a relatively high degree of success. This indicates that CFD models of this type can be accurate and perhaps have not been validated to this extent previously because of the inherent difficulties arising from most real processes.

In recent years, interest in digital watermarking has grown significantly. Indeed, the use of digital watermarking techniques is seen as a promising mean to protect intellectual property rights of digital data and to ensure the authentication of digital data. Thus, a significant research effort has been devoted to the study of practical watermarking systems, in particular for digital images. In this thesis, a practical and principled approach to the problem is adopted. Several aspects of practical watermarking schemes are investigated. First, a power constaint formulation of the problem is presented. Then, a new analysis of quantisation effects on the information rate of digital watermarking scheme is proposed and compared to other approaches suggested in the literature. Subsequently, a new information embedding technique, based on quantisation, is put forward and its performance evaluated. Finally, the influence of image data representation on the performance of practical scheme is studied along with a new representation based on independent component analysis.

Business intelligence (BI) systems comprise one of the largest and fastest growing areas of IT expenditure in companies today. Companies’ experiences with deriving benefits from these systems are still mixed. One of the differences between BI and other types of information systems is that how BI systems are used, not just whether they are used, can have a major impact on the benefits derived. Therefore the characteristics of BI users and the organizations within which they work can have a disproportionate impact on the benefits derived from investments in BI. Organizational competence is one way to evaluate the characteristics of individuals and organizations relative to their ability to achieve organizational goals. This dissertation examines the characteristics of BI users and their organizations within the framework of organizational competences. Models representing those competences at both the individual and organizational level are presented. A combined competency model and resulting emerging competences are proposed that, if adopted, can improve the likelihood of organizations realizing benefits from their BI investments.

Information freely generated, widely distributed and openly interpreted is a rich source of creative energy in the digital age that we live in. As we move further into this irrevocable relationship with self-growing and actively proliferating information spaces, we are also finding ourselves overwhelmed, disheartened and powerless in the presence of so much information. We are at a point where, without domain familiarity or expert guidance, sifting through the copious volumes of information to find relevance quickly turns into a mundane task often requiring enormous patience. The realization of accomplishment soon turns into a matter of extensive cognitive load, serendipity or just plain luck. This dissertation describes a theoretical framework to analyze user interactions based on mental representations in a medium where the nature of the problem-solving task emphasizes the interaction between internal task representation and the external problem domain. The framework is established by relating to work in behavioral science, sociology, cognitive science and knowledge engineering, particularly Herbert Simon’s (1957; 1989) notion of satisficing on bounded rationality and Schön’s (1983) reflective model. Mental representations mediate situated actions in our constrained digital environment and provide the opportunity for completing a task. Since assistive aids to guide situated actions reduce complexity in the task environment (Vessey 1991; Pirolli et al. 1999), the framework is used as the foundation for developing mediating structures to express the internal, external and mental representations. Interaction aids superimposed on mediating structures that model thought and action will help to guide the “perpetual novice” (Borgman 1996) through the vast digital information spaces by orchestrating better cognitive fit between the task environment and the task solution. This dissertation presents an ontology centric architecture for mediating interactions is presented in a semantic web based e-commerce environment. The Design Science approach is applied for this purpose. The potential of the framework is illustrated as a functional model by using it to model the hierarchy of tasks in a consumer decision-making process as it applies in an e-commerce setting. Ontologies are used to express the perceptual operations on the external task environment, the intuitive operations on the internal task representation, and the constraint satisfaction and situated actions conforming to reasoning from the cognitive fit. It is maintained that actions themselves cannot be enforced, but when the meaning from mental imagery and the task environment are brought into coordination, it leads to situated actions that change the present situation into one closer to what is desired. To test the usability of the ontologies we use the Web Ontology Language (OWL) to express the semantics of the three representations. We also use OWL to validate the knowledge representations and to make rule-based logical inferences on the ontological semantics. An e-commerce application was also developed to show how effective guidance can be provided by constructing semantically rich target pages from the knowledge manifested in the ontologies.

A research report submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, Johannesburg, in partial fulfilment of the requirements for the degree of Masters of Science, Industrial Engineering, 31 March 2015 / The purpose of this research was to determine whether principles promoted by a programme management discipline be defined as critical success factors (CSFs) for the realisation of post-implementation benefits from an ERP investment in an organisation. The research method consists of a single case study with three embedded cases. The main source of data was the use of semi-structured interviews, and business documentation was used to corroborate findings. It was concluded that of the eleven CSFs proposed, three were highly likely CSFs, seven were likely but had certain conditions attached, and one is most likely not a CSF. It was also discovered that four additional CSFs proposed in interviews conducted may be relevant, and that four themes exist within the case data analysed. These may form part of future work.

A Doctoral thesis submitted in fulfilment of the requirement for the degree of Doctor of Philosophy in Information Systems, Division of Information Systems School of Economic and Business Sciences Faculty of Commerce, Law and Management.Johannesburg, March 2017 / This thesis is about the adoption of an Information System (IS) at a country level. Information Systems literature addresses adoption of IS at an individual level, organisational level or national/country level. Each level of analysis has its own complexities. However, literature acknowledging these varied complexities has not been forth coming. That is, literature has more studies done at either individual or organisational, and hardly at national or country level. This thesis argues that the adoption of an information system (also referred to as an innovation) at country level is a multi-dimensional and multi-level phenomenon. Existing literature and previous studies have hardily addressed fully, this complexities and multi-dimensionalism, although it has been noted that countries experience and internalise the innovation adoption, as a social process, differently. The study was on a developing country adopting a Physical Address System (PAS), herein seen as an IS innovation. In this thesis, PAS is seen as a social system comprising of artefacts (digital and visual representations), physical world, residents and organisations as stakeholders. The goal of the study was to conceptualise a multi-dimensional framework for adopting a Physical Address System, in the context of a developing country. Since the thesis argument is that the adoption of IS at a country level is even more complex, varied theories were employed as lenses to tackle the various aspect of the study. These lenses are the Diffusion of Innovation, the Stakeholder Theory, Upper Echelon Theory and the Contextualist Approach. Following the interpretivist philosophy, a case study was employed as a research strategy, using Botswana as a developing country case. The research design included semi-structured interviews with stakeholders, observations, policy documents. The data was analysed, discussed, synthesised and interpreted using thematic framework analysis method. Informed by the empirical evidence and the existing literature, this thesis conceptualises that the adoption of the Physical Address System ought to be done sensitive to the developing country as a multi-dimensional social system. This multi-dimensional social system includes the roles of stakeholders, determinants of innovation and context. The contribution of the thesis is in four folds; theoretical, methodological, practical, and contextual. Theoretically, the thesis conceptualised a multi-dimensional framework for the adoption of the Physical Address System in a developing country. Methodologically, the thesis contributed by following an interpretive philosophy and a case study as appropriate for understanding the complexities of adopting an information system, employing a case. Practically, the thesis, through the framework, may inform practitioners with ways to adopt a physical address system. Contextually, the thesis gives insight into the uniqueness of a developing country adopting an information system. Keywords: Developing Country, Adoption, Physical Address System, Stakeholder Theory, Upper Echelon Theory, Diffusion of Innovation, Context / GR2018

A research report submitted to the Faculty of Commerce, University of the Witwatersrand, Johannesburg, in partial fulfilment of the requirements for the degree of Master of Commerce, 1998 / Corporate Governance requires management to report to its stakeholders on Internal Control Systems. Corporate Governance is the system through which organisations are directed and controlled. To meet these requirements management needs a mechanism through which they can stay abreast of such control systems. The aim of this research is to evaluate whether such a mechanism can be provided for Information System Development (ISD) projects, through Control Self-Assessment questionnaire and / or workshop techniques. [Abbreviated abstract. Open document to view full version} / MT2017

This work introduces a novel inversion-based neurocontroller for solving control problems involving uncertain nonlinear systems which could also compensate for multi-valued systems. The approach uses recent developments in neural networks, especially in the context of modelling statistical distributions, which are applied to forward and inverse plant models. Provided that certain conditions are met, an estimate of the intrinsic uncertainty for the outputs of neural networks can be obtained using the statistical properties of networks. More generally, multicomponent distributions can be modelled by the mixture density network. Based on importance sampling from these distributions a novel robust inverse control approach is obtained. This importance sampling provides a structured and principled approach to constrain the complexity of the search space for the ideal control law. The developed methodology circumvents the dynamic programming problem by using the predicted neural network uncertainty to localise the possible control solutions to consider. Convergence of the output error for the proposed control method is verified by using a Lyapunov function. Several simulation examples are provided to demonstrate the efficiency of the developed control method. The manner in which such a method is extended to nonlinear multi-variable systems with different delays between the input-output pairs is considered and demonstrated through simulation examples.