In recent years, such vast archives of video information have become available that human annotation of content is no longer feasible; automation of video content analysis is therefore highly desirable. The recognition of semantic content in images is a problem that relies on prior knowledge and learnt information and that, to date, has only been partially solved. Salient analysis, on the other hand, is statistically based and highlights regions that are distinct from their surroundings, while also being scalable and repeatable. The arrangement of salient information into hierarchical tree structures in the spatial and temporal domains forms an important step to bridge the semantic salient gap. Salient regions are identified using region analysis, rank ordered and documented in a tree for further analysis. A structure of this kind contains all the information in the original video and forms an intermediary between video processing and video understanding, transforming video analysis to a syntactic database analysis problem. This contribution demonstrates the formulation of spatio-temporal salient trees the syntax to index them, and provides an interface for higher level cognition in machine vision.

Doctor of Philosophy / Department of Educational Leadership / W. Franklin Spikes / The purpose of this study was to identify the informational resources used by small manufacturing owners who were in the NAICS 336 sector of manufacturers in Sedgwick County, Kansas, and determine which are most helpful to the owner in managing their business. Prior studies have utilized all small businesses and have not examined industry specific information gathering. A qualitative method was used in the study using a semi-structured interview process to provide in-depth information from the perspective of the small manufacturing owner. The semi-structured interview identified the information resources used by small manufacturing owners in managing their business and the level of helpfulness those resources had for them. Ten small manufacturers whose owner was on site and employed from 20-100 employees were used in the study. Informal resources were most often used by this group of manufacturers. The Internet was identified by the manufacturers as the most important and useful resource to managing their businesses. The small manufacturers used multiple resources to manage their business and rarely relied on only one resource of information whether it was for financial, management, marketing/sales or benchmarking information. Formal sources of information such as colleges, trade schools and seminars were less frequently used. Nonformal resources which were mostly governmental and included the Small Business Association, Small Business Development Center, Manufacturing Extension Partnerships, SCORE, local and federal government resources were rarely used by this group of manufacturers.

M.Ing. / Failure is a part of every day life. Information systems project failures are reported regularly in the media. The growth of the information technology industry in its entirety has been very dramatic over the last decade. Large information systems projects have always posed as a challenge for development teams and organizations. The nature of these challenges have increased with the growth of information systems projects over the last decade. With the critical role that information systems have in organizations, it is critical that information systems are delivered on time within budget and according to the requirements. Unfortunately many of the information systems projects that have been started, are cancelled, not on time or over budget. It is the purpose of this dissertation to discuss some of the managerial causes of why information systems projects fail, and also to address issues that could prevent the failure of information systems projects.

This dissertation concerns the evaluation of the utilisation of e-Learning in public Further Education and Training Colleges for the provision of education and training. The literature study investigates the concept e-Learning, related terminology, benefits and drawbacks and their relevance to the FET sector in South Africa. The study identifies basic e-Learning requirements, that is, infrastructure, technical and human resources for the implementation of e-Learning in FET Colleges. Government policy initiatives that seek to assist FET Colleges in the use of Information Communication Technology (ICT), challenges and opportunities for the provision of education and training that FET Colleges have in e-Learning are also investigated. In order to realise the objectives of the study, separate structured pre-coded interviews with three senior managers representing the following Further Education and Training Colleges were held, Central Johannesburg, Ekurhuleni West and Sedibeng. The interviews covered issues relating to policy and availability of infrastructure and other resources for e-Learning. Self-administered questionnaires were also used to capture the views of middle managers on various e-Learning issues. Findings indicate that e-Learning is a superficial matter for FET institutions – a little known phenomenon. The available ICT infrastructure in FET Colleges is mainly used for management and administration. Access to the Internet is mainly for managers and educators and very limited for learners. There are also conflicting views on e- Learning between senior management and middle managers due to miscommunication. There is a skills shortage as far as e-Learning is concerned. / Mr. H.B. Klopper

Ph.D. (Energy Studies) / The underlying assumption of this thesis is that information is critical for effective energy policy support. The objectives set and attained for this study were to: • assess existing solutions to the problem of supplying energy information; • consider developments in integrated conceptual solutions to the problems; • identify and analyse data structures for providing a single integrated information system as a solution; • evaluate and identify local energy and economic data sources for input to the proposed system; • provide a framework for information outputs in such an integrated system; and • make recommendations on the practical implementation of the conceptual . solution. The concepts underlying the basic assumption, and other parameters were clarified in Chapter 1. In particular, the link between policy formulation, modeling, and data collection, management and dissemination was shown. A study and assessment of existing energy information systems was then carried out. This investigated national systems in South Africa and other countries, and the systems of three international organisations. The structure and management of these energy information systems was assessed, as was the degree to which each provided the necessary support for policy formulation. Legislation was, where applicable, also considered. It was determined from this study and assessment that there were no integrated information systems to provide adequate information for energy policy makers through a single access point. An integrated single access point system was postulated at the conceptual level in Chapter 3. The major contribution to this conceptualisation is the concept of a single standardised referential data set A single referential data set can be used to classify all data at every level of the proposed integrated national energy information system. This integrated approach supports policy formulation and allows for the most efficient use of national resources in maintaining information. The proposed system was considered as a data flow model, to provide an overview of the collection, processing and dissemination of data. Next, a system model was developed. This divided the system into two broad divisions: the central control functions, including the referential data set; and the data modules, which provide the structure for capturing and maintaining the collected data. The final stage in this chapter . was the development of a data model which shows the relationships between the elements of the system model, particularly the relationship between the referential data set and all other data. In Chapter 4, this data model was developed further to identify entity sets and their content descriptions. It indicates in tabular format the structure of the different system components, and then shows diagrammatically the high level linkages between these components; This analysis of the proposed system is the precursor to work by systems analysts who are responsible for the final design specifications for system software. In Chapter 5 the objective was to determine sources of data for input to the national system. The South African energy economy was briefly studied and assessed for both supply and demand side information. Finally in Chapter 6, a set of output templates was developed. Once fully implemented, these will provide system users with all the necessary information to support policy formulation. Desired outputs from a system must be known before the system is designed, and are therefor included.

Ph.D. / The present study originated from a realisation about the unique nature of the medical domain and about the limitations of existing risk-management methodologies with respect to incorporating the special demands and salient features of the said domain. A further incentive for the study was the long-felt need for proper Information Technology (IT) risk management for medical domains, especially in the light of the fact that IT is playing an ever-greater part in the rendering of health-care services. This part, however, introduces new information-security challenges every day, especially as far as securing sensitive medical information and ensuring patients' privacy are concerned. The study is, therefore, principally aimed at making a contribution to improving IT risk management in the medical domain and, for this reason, culminates in an IT risk-management model specifically developed for and propounded in the medical domain. While developing this model, special care was taken not only to take into consideration the special demands of the said domain when assessing IT risks but also that it would be suited to the concepts, terminology and standards used in and applied to this domain every day. The most important objectives of the study can be summarised as follows: A thorough investigation into modern trends in information security in the medical domain will soon uncover the key role IT is playing in this domain. Regrettably, however, this very trend also triggers a steep increase in IT riskincidence figures, which, in this domain, could often constitute the difference between life and death. The clamant need for effective risk-management methods to enhance the information security of medical institutions is, therefore, self-evident. After having explored the dynamic nature of the medical domain, the requirements were identified for a risk-management model aimed at effectively vi managing the IT risks to be incurred in a typical medical institution. Next, a critical evaluation of current risk-assessment techniques revealed that a fresh approach to IT risk management in medical domains is urgently necessary. An IT risk-management model, entitled "RiMaHCoF" (that is, "Risk Management in Health Care — using Cognitive Fuzzy techniques"), was developed and propounded specifically for the medical domain hereafter. The proposed model enhances IT risk management in the said domain in the sense that it proceeds on the assumption that the patient and his/her medical information constitute the primary assets of the medical institution.

D.Litt. et Phil. / This research addresses the concept of “information architecture” as a way of visualising and describing the various information assets and interaction of these assets within the organisation. The research further provides definitions of information and contextualises the information audit and information model as key tool for the information manager in establishing the information architecture. The establishment of the information architecture is illustrated through a case study within a large conglomeration of companies requiring a scaleable information architecture in order to address its information requirements. Through a process of interviews the information requirements of key decisions makers are established. These requirements are translated into an information architecture that forms the basis of support in delivering future information requirements for the company. The results of this research have been successfully implemented and now form the foundation of all future information management projects within this particular case study.

Ph.D. (Leadership Performance and Change) / Organisations worldwide have made substantial investments in the training of their employees even though it is generally understood that only a small amount of learning is usually transferred to the work environment. The knowledge of the extent of employer investments in training compared to the benefits that accrue from this training for the organisation underlies the continuous debate concerning the return on investment and the effectiveness of the transfer of learning. Learning Transfer is defined as the application of knowledge, skills and attitudes learned from training and the subsequent maintenance of it over a period of time. This paucity of learning transfer, in spite of the substantial investment in training by employers, must be considered against the reality of organisations continuously being confronted by demands emanating from developments such as globalisation. Equally, technological developments necessitate change in the nature of work and consequently in the knowledge and skills required by employees to perform the work, as well as for organisations to remain globally competitive. The aim of this study is to determine whether a causal relationship exists between management support for learning and the effectiveness of learning transfer. In pursuance of finding causes for the lack of learning transfer, the study has the further objective of determining whether management support could be elevated above other differentiators, such as motivation of the trainee, training design and the workplace or organisational climate factors. The study is premised on the perception that management exercises a great deal of influence over their employees and that they determine organisational outcomes because of their decision-making authority. This gave rise to the presumption that management plays a leading role in ensuring that effective learning transfer is achieved and that the nature and extent of management‟s influence determines whether effective and efficient learning transfer is realised.

The main purpose of this project is to investigate the state of the art of the Algerian National Archives (ANA) so as to design an automated system that responds to the needs of this institution. The concept of archives is defined. The effects of computerisation on archives are investigated. Some automated archives systems, around the world, are examined. The various obstacles impeding the development of a technology capable of processing Arabic script are reviewed. Some solutions are also discussed. The case of Algeria in the context of the Arab world is taken as en example. A number of problems hampering the transfer of technology are identified. The study is concentrated on the state of the art of the ANA. It is carried out using a variety of data collection techniques; including questionnaires, interviews, observation and the author's own experience of the ANA. Severel problem areas are identified; including: the lack of resource sharing between the various institutions and the access speed to documents, to name just a few. The objectives of the system to be designed and implemented ere identified. A prototype user-friendly system, using the Query language, dBASE III PLUS and Clipper, is developed to simulate some of the various tasks carried out within records management institutions. An explanation of how can the system be operated is provided. An evaluation of the prototype system is carried out. A number of recommendations to improve the system are presented. Amongst these, is the necessity to provide a much faster system. The prototype system is thus redeveloped using, this time, Turbo Pascal. Apart from speed, no alterations or additions are introduced to the one developed using dBASE. Because the project has not been taken to the last stages of the System Development Life Cycle (SDLC), a number of recommendations are made regarding the steps that the ANA should observe to ensure a smooth system implementation and maintenance.

Social media offers researchers vast amount of unstructured text as a source to discover hidden knowledge and insights. However, social media poses new challenges to text mining and knowledge discovery due to its short length, temporal nature and informal language. In order to identify the main requirements for analysing unstructured text in social media, this research takes a case study of a large discussion forum in the diabetes domain. It then reviews and evaluates existing text mining methods for the requirements to analyse such a domain. Using domain background knowledge to bridge the semantic gap in traditional text mining methods was identified as a key requirement for analysing text in discussion forums. Existing ontology engineering methodologies encounter difficulties in deriving suitable domain knowledge with the appropriate breadth and depth in domain-specific concepts with a rich relationships structure. These limitations usually originate from a reliance on human domain experts. This research developed a novel semantic text mining method. It can identify the concepts and topics being discussed, the strength of the relationships between them and then display the emergent knowledge from a discussion forum. The derived method has a modular design that consists of three main components: The Ontology building Process, Semantic Annotation and Topic Identification, and Visualisation Tools. The ontology building process generates domain ontology quickly with little need for domain experts. The topic identification component utilises a hybrid system of domain ontology and a general knowledge base for text enrichment and annotation, while the visualisation methods of dynamic tag clouds and cooccurrence network for pattern discovery enable a flexible visualisation of these results and can help uncover hidden knowledge. Application of the derived text mining method within the case study helped identify trending topics in the forum and how they change over time. The derived method performed better in semantic annotation of the text compared to the other systems evaluated. The new text mining method appears to be “generalisable” to other domains than diabetes. Future study needs to confirm this ability and to evaluate its applicability to other types of social media text sources.