Search

Using the Internet for teaching and learning has become a trend in modern higher education, facilitated through the exploitation of advanced computing technologies. Virtual Learning Environment (VLE) applications support online learning over the Internet, and VLEs have thus emerged as e-learning domains that are essential prerequisites in cutting edge design and implementation technologies in education. Service Oriented Computing (SOC), as a novel software development and implementation approach, has become an active area of research and development. Web services, as an example of SOC, support the integration of software applications in an incremental way, using existing platforms and languages that utilize and adopt existing legacy systems. Thus, VLEs should be particularly well suited to Web services through the SOC approach. VLE services is a field subjected to continuous development but VLEs as Web services are still not generally accessible for academic institutions, although they have been adopted by some scientific projects. The next generation of VLEs should address the limitations of the current online systems by providing a richer context for online learning, one that is sensitive to the specific domain requirements of e-learning. Web Services Matching and Selection (WSMS), as a part of the functional requirements of Web services, has received less attention from SOC researchers. It involves discovering a set of semantically equivalent services by filtering a set of available services based on service metadata, and instantaneously selecting the best possible service. WSMS is the discovery of a service by a user, where correspondence is established between the objectives of the consumer and the capabilities of the service. It thereby aims to match and select the optimal service that best meets the requestor's needs. The main aim of this doctoral work is to explore novel architectural designs for VLEs, based on the SOC paradigm and its related techniques. In addition, this investigation aims to extend the core ideas behind VLE tools, which are gradually becoming dominant within academic institutes. Another aim is to devise a policy- based technique to enforce security requirements for VLEs and to build a test-bed for VLE security based on Modular Moodle. The fundamental contribution of this thesis that it demonstrates that VLEs can be considered as services, which can be published, discovered and composed as perceived in the SOC paradigm. An additional contribution to the knowledge is that it has built a new extension to the structure of Web services: the Web Services Matching and Selection (WSMS) system. Another contribution to the knowledge is that traditional security requirements have been modified to cater for the highly mobile and changeable environment of VLEs; this has been achieved through policy- based techniques. These contributions to the body of knowledge have been published in learned journals and at conferences.

People from different cultures use web interface in different ways, expect different visual representation, navigation, interaction, mental model, and layouts etc., and have different communication patterns and expectation. In the context of globalisation, web localisation becomes a powerful strategy to acquire an audience in a global market. Therefore, web developers and designers have to make adaptations to fit the needs of people from different cultures, and the way cultural factors are integrated into web interface design needs to be improved. Most previous research lacks an appropriate way to apply cultural factors into web development. No empirical study of the web interface has been carried out to support the cross-cultural web design model. It is noted that no single model can support all cross-cultural web communication but a new model is needed to bridge the gap and improve the limitations. Thus the research aim was to build a new model of cross-cultural web design to contribute to effective communication. Following an extensive literature review, a local web audit was conducted, then a series of experiments with users to gather and evaluate data and build and validate the new model. A new model, based on a study of British and Taiwanese users, was formulated and validated demonstrating that content and message remain the core of web design but the performance of the selected users is influenced by the cultural dimension and cultural preferences and this, in turn impacts on the effectiveness of the web communication. For the British user sample, ease of using the website was seen to be strongly related to desirability. Taiwanese users showed preference for visual pleasure but no relationship between efficient performance and desirability. The resultant model contributes to the knowledge of how to design effective web interfaces for British and Taiwanese cultures and is replicable for the purpose of comparing approaches to designing for other cultures.

As the web expands exponentially, the need to put some order to its content becomes apparent. Hypertext categorization, that is the automatic classification of web documents into predefined classes, came to elevate humans from that task. The extra information available in a hypertext document poses new challenges for automatic categorization. HTML tags and linked neighbourhood all provide rich information for hypertext categorization that is no available in traditional text classification.

Adaptive hypermedia has provided a way information can be presented online. Based on adaptive presentation and adaptive navigational support, a static page can now be dynamically personalised for an individual user. Users who possess different needs, interests and background knowledge can now be provided with a different presentation of the same information. Many frameworks for adaptive hypermedia systems and applications have been proposed that use different strategies. This thesis proposes a new approach for the presentation and personalisation of links based on the idea of a multi-dimensional linkbase. It is the notion that describes a single linkbase that contains links annotated with metadata that place the links in several different contextual dimensions at once. These sets of links signify different dimensions of expertise of the user and are encoded to condition the visibility of links. This work builds upon the implementation of FOHM and Auld Linky at Southampton University. To provide users with control over the personalisation of their links, the users are provided with navigational tools for the presentation of these links. The presentation of the links depends on the preferences of the users and the linkbases they have enabled and disabled. This facilitates flexibility and reduces the user syndrome of ‘too many-irrelevant-additional links’. Four straightforward adaptive systems have been developed to demonstrate the diversity of the link service approach, and in particular the concept of a multi¬dimensional linkbase, which has been applied into a Web-based prototype, an inquiry-led personalised navigation system. This thesis also documents the formal evaluation studies undertaken, which demonstrates that such a proposal is practicable and meaningful to a user.

Real-time applications such as Voice-over-IP, streaming multimedia, multi-player gaming and pervasive virtual environments account for a massive proportion of the Internet economy. These services consist of traffic flows which have requirements in terms of bandwidth, loss, delay and jitter in order to provide a usable experience to an end-user. Current approaches to providing specialised service for real-time traffic involve queueing mechanisms, resource reservation, traffic engineering or admission control. These strategies all either have scalability issues, do not work in an end-to-end fashion, or are manual and off-line in nature. This dissertation describes and evaluates new approaches for on-line routing of real-time flows. Our framework for this work is the Cognitive Packet Network (CPN), a self-aware routing protocol which uses a user-directed, goal-based reinforcement learning technique to route flows according to the metrics which most impact their performance. We first compare CPN with an industry standard routing protocol (OSPF), demonstrating through experiments with multiple flows carrying real-time traffic that CPN can quickly converge to the optimal route in a network, and that the speed with which this occurs is proportional to its exploratory overhead. This work is the first. to explore CPN's performance with multiple flows competing for the available network resources. This leads to our second contribution where we investigate the impact of competition among selfish network flows and the potential for routing oscillations. We demonstrate that. these oscillations can harm overall QoS, and develop methods for eliminating them without sacrificing the benefits of CPN's exploratory nature. Towards supporting applications which have diverse Quality of Service (QoS) requirements, we develop goals in terms of two metrics (loss and delay) and show that CPN can adaptively learn paths in terms of each; in cases of moderate congestion it can decrease network losses by up to a factor of 10. Finally, we show that by using an overlay network, the benefits of CPN routing can be introduced into existing networks with low overhead and without modifying their underlying routing mechanisms.

Although, today, website design still contains many variable design elements which strongly influence its design outcome, like browser usage, platforms, user preferences, window size, connection speed, computer speed, colour support, software support, and font support, the original design idea still plays an important role. Emotion is difficult to measure or evaluate. Designers need to understand its importance and how to measure emotion efficiently in order to undertake the subsequent design stage. Positive emotions are as important as negative ones-positive emotions are critical to learning, curiosity, and creative thought, and today research is turning toward this dimension (Norman, 2004b: 19). To understand users' negative emotions can supply a better solution for design problems. Both users' likes and dislikes are important when exploring the role of emotional design. The fieldwork consisted of two pilot tests (the initial pilot test-I. P. T. and the second pilot test-S. P. T. ), a website impression test (W. I. T. ), and an improved website impression test (I. W. I. T. ). The research methods involved Think-Aloud usability testing, Emotional Probes testing (EPs testing), and questionnaires. In addition, the evaluation method used a coding system and statistics. Triangulation was used in making consistent the process methods and evaluation methods. The research outcomes include a fashion website design reference model, an insurance website design reference model, and the website impression test system. Through conducting the fieldwork, the importance of emotional design can be identified in the website design field. The whole research process and research outcomes supply one solution for researchers and designers to predict users' emotions in design outcomes.

A rapidly emerging need within virtual environments is the ability to accommodate a large, distributed user-base. In particular, Internet-based distributed virtual environments represent a challenge in providing a heterogeneous environment accessible by many thousands of users. Interest Management represents the process of filtering network data based on relevance to users, in order to make the most efficient use possible of available network capacity. In this thesis, the potential for a visual-attention based refinement to interest management is explored, which utilises a spotlight-based model of human attention. By considering the relationship between attention and interest management, this thesis illustrates a tendency within existing systems to use a simplistic or object-based approach to measure relevance, typically based on proximity to the user or visibility. By considering the current state of the art alongside common theories of visual attention, it is hypothesised that a refinement aiming to accommodate a spotlight model may offer an approach capable of more efficient use of bandwidth when compared to existing techniques. In particular, it is suggested a spotlight model would be capable of accommodating key aspects of human visual attention in a processor-efficient and composable manner. Bandwidth conserved by the use of such an approach may be subsequently re-utilised to provide a higher degree of interactivity within a distributed virtual environment, or to support greater numbers of simultaneous users. In order to evaluate the effectiveness of a spotlight-based approach, a test environment is created which allows users to participate in a number of activities alongside a large volume of simulated clients. Throughout these activities, interest management is changed seamlessly between variations of a spotlight approach, extremes, and an approximation of the current state of the art. Results obtained from a total of 15 users indicate a preference for such an approach when compared to the current state of the art in 80% of subjects, and suggest the capacity to reduce available bandwidth by up to 60%, dependent on task, without perceivable impact.

We live in a world where we are surrounded by ever increasing numbers of images. More often than not, these images have very little metadata by which they can be indexed and searched. In order to avoid information overload, techniques need to be developed to enable these image collections to be searched by their content. Much of the previous work on image retrieval has used global features such as colour and texture to describe the content of the image. However, these global features are insufficient to accurately describe the image content when different parts of the image have different characteristics. This thesis initially discusses how this problem can be circumvented by using salient interest regions to select the areas of the image that are most interesting and generate local descriptors to describe the image characteristics in that region. The thesis discusses a number of different saliency detectors that are suitable for robust retrieval purposes and performs a comparison between a number of these region detectors. The thesis then discusses how salient regions can be used for image retrieval using a number of techniques, but most importantly, two techniques inspired from the field of textual information retrieval. Using these robust retrieval techniques, a new paradigm in image retrieval is discussed, whereby the retrieval takes place on a mobile device using a query image captured by a built-in camera. This paradigm is demonstrated in the context of an art gallery, in which the device can be used to find more information about particular images. The final chapter of the thesis discusses some approaches to bridging the semantic gap in image retrieval. The chapter explores ways in which un-annotated image collections can be searched by keyword. Two techniques are discussed; the first explicitly attempts to automatically annotate the un-annotated images so that the automatically applied annotations can be used for searching. The second approach does not try to explicitly annotate images, but rather, through the use of linear algebra, it attempts to create a semantic space in which images and keywords are positioned such that images are close to the keywords that represent them within the space.

The research reported in this thesis assesses the impact of media quality degradations in Internet multimedia conferencing on users. Low quality audio and video can be experienced, therefore it is important to determine the minimum levels of quality needed to perform specific tasks. This has most commonly been investigated using subjective measures, however the research reported in this thesis adopted a 3-factor evaluation framework of task performance, user satisfaction and user cost. User satisfaction was measured subjectively, whereas physiological indicators of perceptual strain were utilised to measure user cost. Physiological measures provide continuous data throughout a session, are not subject to cognitive mediation and taking such measurements does not interfere with the user's task. Five experiments were performed investigating audio and video quality degradations. With the exception of one passive listening task, all tasks used were based on remote interviews, as they fully exploit the capabilities of the application. Results showed that physiological responses to media quality degradations can be detected in passive, perceptual tasks. However, active participation in a task made it more difficult to detect changes due to quality degradations. In all experiments physiological measures gave information on the nature of the tasks being performed and effects of variables such as order. The results of this research were then used in three further experiments in the areas of VR and web quality of service and design. In conclusion, the physiological measures utilised in the research reported in this thesis can be employed to assess the impact of media quality degradations in passive perceptual tasks and to give general information about the nature of the task being performed.

The central purpose of this thesis is to investigate whether users have distinct preferences for specific navigational patterns in multimedia: that is preferences for moving through multimedia. Subsidiary questions are whether users have preferences for working strategies, (the mental approach to investigating software) whether these preferences are similar for specific groups and whether these preferences are affected by the software's system and navigational design. Four groups were investigated within two ranges: children to adults, and novices to experts. The literature review revealed four different perspectives of investigating navigation: user, designer, pedagogy and human computer interaction and although this research concentrates on the first two perspectives the other two are integral and of equal importance. Two empirical studies elicited the navigational information. The first studied pairs of children undertaking set tasks in multimedia, and demonstrated that although each pair had definite preferences, each group did not utilise the full pattern range discerned from the observations, literature review and multimedia package analysis. The second study was redesigned using individual adults to ascertain the full range of preferred patterns in use. The essential element from the investigations was the wide range of variation between individuals and within groups. There was a gradual progression in their range and speed using these patterns, related to their skills, abilities and experience, and each individual could be placed along a continuum. Topologies of the multimedia packages and diagrams of the fit of the navigation patterns were included. Finally an expert panel was convened to verify the pattern range and their comments supported the new classification. The research outcomes included navigational patterns and working strategies classifications, future techniques for designers, and user methods. These will create more successful and informed multimedia, and forward developments and improvements in the design of high quality user preference software.

Digital media cultural practices continue to address the social, cultural and aesthetic contexts of the global information economy, perhaps better called ecology, by inventing new methods and genres that encourage interactive engagement, collaboration, exploration and learning. The theoretical framework for creative critical technology evolved from the confluence of the arts, human computer interaction, and critical theories of technology. Molding this nascent theoretical framework from these seemingly disparate disciplines was a reflexive process where the influence of each component on each other spiraled into the theory and practice as illustrated through the Constructed Narratives project. Research that evolves from an arts perspective encourages experimental processes of making as a method for defining research principles. The traditional reductionist approach to research requires that all confounding variables are eliminated or silenced using methods of statistics. However, that noise in the data, those confounding variables provide the rich context, media, and processes by which creative practices thrive. As research in the arts gains recognition for its contributions of new knowledge, the traditional reductive practice in search of general principles will be respectfully joined by methodologies for defining living principles that celebrate and build from the confounding variables, the data noise. The movement to develop research methodologies from the noisy edges of human interaction have been explored in the research and practices of ludic design and ambiguity (Gaver, 2003); affective gap (Sengers et al., 2005b; 2006); embodied interaction (Dourish, 2001); the felt life (McCarthy & Wright, 2004); and reflective HCI (Dourish, et al., 2004). The theory of critical creative technology examines the relationships between critical theories of technology, society and aesthetics, information technologies and contemporary practices in interaction design and creative digital media. The theory of critical creative technology is aligned with theories and practices in social navigation (Dourish, 1999) and community-based interactive systems (Stathis, 1999) in the development of smart appliances and network systems that support people in engaging in social activities, promoting communication and enhancing the potential for learning in a community-based environment. The theory of critical creative technology amends these community-based and collaborative design theories by emphasizing methods to facilitate face-to-face dialogical interaction when the exchange of ideas, observations, dreams, concerns, and celebrations may be silenced by societal norms about how to engage others in public spaces. The Constructed Narratives project is an experiment in the design of a critical creative technology that emphasizes the collaborative construction of new knowledge about one's lived world through computer-supported collaborative play (CSCP). To construct is to creatively invent one's world by engaging in creative decision-making, problem solving and acts of negotiation. The metaphor of construction is used to demonstrate how a simple artefact - a building block - can provide an interactive platform to support discourse between collaborating participants. The technical goal for this project was the development of a software and hardware platform for the design of critical creative technology applications that can process a dynamic flow of logistical and profile data from multiple users to be used in applications that facilitate dialogue between people in a real-time playful interactive experience.

Hypertext is a multi-linear electronic, textual and interactive environment to present information. The objective of such an environment is that readers may browse through linked, cross-referenced, annotated texts in a multi-sequential manner, and thus, it is believed, to improve the learning. However, early and current research findings have revealed some mixed results concerning the alleged advantage of hypertext on learning over paper-based documents. Researchers have identified the lack of research about the cognitive processes and the strategies that readers use during reading as one of the main factors for such results. As a result, there is a need and scope for further research in modelling the cognitive processes involved in reading comprehension and the reading strategies in a hypertext environment. This research addresses some of the gaps in the field by proposing a model that represents the sequence of events that take place during reading in a Web-based hypertext environment. Also, emphasis is placed on the strategies that readers use during hypertext reading and on the potential effect of different reading goals on reading comprehension. The evaluation of the model and the other hypotheses is conducted in two experiments using qualitative and quantitative methods. The first experiment employs the think aloud method. Forty two subjects participated. The results demonstrated that the proposed model precisely describes the sequence of events that take place during hypertext reading. They did not reveal any significant difference between different reading goals and understanding. They revealed four reading strategies: serial, serial overview, mixed, and mixed overview, and they identified three factors that influence the selection of hyperlinks: coherence, link location, and personal interest. The second experiment is an independent samples design experiment with ninety subjects. The results confirmed those found in the first experiment. The current study makes a contribution in the field of hypertext reading by proposing and evaluating a procedural model and by making this model graphic. By doing so it addresses some of the voids in the field, expands our understanding of the reading processes and the reading strategies, and provides practical guideliness which are enhanced to promote design supporting effective learning processes.

Recognising the World Wide Web as both a disruptive and pervasive technology, in tandem with the emergence of devices with widely ranging capabilities through which it is navigated, the role of the Web as an information system in pervasive computing is becoming an important concern. A key characteristic of the Web is the ability to access distributed resources, navigating between them by following links. However, this relatively static interaction model is not suitable in scenarios where resources are not at 'well known' locations, for example when copies or versions exist locally. This thesis concerns an augmentation of the Web to provide access to spontaneously available local resources, enriched with open hypermedia linking, for local participants in an impromptu network. The approach taken has been to identify scenarios that serve to scope the application space and then analyse and develop, through a series of prototype experiments, different enabling infrastructures for hypermedia link services, grounded on the notion of a framework of cooperating components.

Distributed real-time collaboration, such as group-to-group videoconferencing, is becoming increasingly popular. However, this form of collaboration tends to be less effective than co-located interactions and there is a significant body of research that has sought to improve the collaboration technology through a variety of methods. Some of this research has focused on adding annotations that explicitly represent events that take place during the course of a collaboration session. While this approach shows promise, existing work has in general lacked high-level semantics, which limits the scope for automated processing of these annotations. Furthermore, the systems tend not to work in real-time and therefore only provide benefit during the replay of recorded sessions. The systems also often require significant effort from the session participants to create the annotations. This thesis presents a general-purpose framework and proof of concept implementation for the automated, real-time annotation of live collaboration sessions. It uses technologies from the Semantic Web to introduce machine-processable semantics. This enables inference to be used to automatically generate annotations by inferring high-level events from basic events captured during collaboration sessions. Furthermore, the semantic approach allows the framework to support a high level of interoperability, reuse and extensibility. The real-time nature of the framework means that the annotations can be displayed to meeting participants dUling a live session, which means that they can directly be of benefit during the session as well as being archived for later indexing and replay of a session recording. The semantic annotations are authored in RDF (Resource Description Framework) and are compliant to an OWL (Web Ontology Language) ontology. Both these languages are World Wide Web Consortium (W3C) recommendations. The framework uses rule-based inference combined with knowledge from an external triplestore to generate the annotations. A shared buffer called a tuple space is used for sharing these annotations between distributed sites. The proof of concept implementation uses existing Access Grid videoconferencing technology as an example application domain, to which speaker identification and participant tracking are added as examples of semantic annotations.

Open hypermedia emphasises the separation of hypermedia structure and data, typically achieved through the removal of hypermedia mark-up within data. Such removal enables re-application of the same hypermedia structure to different sets of data. However, the internal organisation of hypermedia structure does not follow these same separation rules as most hypermedia systems continue to embed mark-up within hypermedia structure. This violates the principles of open hypermedia since the objects and connections of a hypermedia structure cannot be re-used within other hypermedia structures. Hence only entire hypermedia structure re-use is possible. This is not only an inefficient use of resources, but it highlights the anomaly that the objects and connections of open hypermedia structure are not actually open to re-use. This thesis describes the Object and Connection Space (OCS) data model. It opens up hypermedia structure by separating the functional and connectional aspects of hypermedia structure. Once assigned to Object and Connection Spaces the productive re-use of hypermedia structure can begin, either as individual hypermedia objects or as collections of selected hyperstructure segments. The thesis also explores how the OCS data model, via structure re-use, enables improved structure versioning, in particular the prevention of revision proliferation. And it considers how the OCS data model benefits the overall maintenance of hypermedia structure by offering a more logical approach to the repair of broken internal routes within hypermedia structure.

This work systematically investigates when and how combining speech output and visual text may facilitate processing and comprehension of sentences. It is proposed that a redundant multimodal presentation of speech and text has the potential for improving sentence processing but also for severely disrupting it. The effectiveness of the presentation is assumed to depend on the linguistic complexity of the sentence, the memory demands incurred by the selected multimodal configuration and the characteristics of the user. The thesis employs both theoretical and empirical methods to examine this claim. At the theoretical front, the research makes explicit features of multimodal sentence presentation and of structures and processes involved in multimodal language processing. Two entities are presented: a multimodal design space (MMDS) and a multimodal user model (MMUM). The dimensions of the MMDS include aspects of (i) the sentence (linguistic complexity, c.f., Gibson, 1991), (ii) the presentation (configurations of media), and (iii) user cost (a function of the first two dimensions). The second entity, the MMUM, is a cognitive model of the user. The MMUM attempts to characterise the cognitive structures and processes underlying multimodal language processing, including the supervisory attentional mechanisms that coordinate the processing of language in parallel modalities. The model includes an account of individual differences in verbal working memory (WM) capacity (c.f. Just and Carpenter, 1992) and can predict the variation in the cognitive cost experienced by the user when presented with different contents in a variety of multimodal configurations. The work attempts to validate through 3 controlled studies with users the central propositions of the MMUM. The experimental findings indicate the validity of some features of the MMUM but also the need for further refinement. Overall, they suggest that a durable text may reduce the processing cost of demanding sentences delivered by speech, whereas adding speech to such sentences when presented visually increases processing cost. Speech can be added to various visual forms of text only if the linguistic complexity of the sentence imposes a low to moderate load on the user. These conclusions are translated to a set of guidelines for effective multimodal presentation of sentences. A final study then examines the validity of some of these guidelines in an applied setting. Results highlight the need for an enhanced experimental control. However, they also demonstrate that the approach used in this research can validate specific assumptions regarding the relationship between cognitive cost, sentence complexity and multimodal configuration aspects and thereby to inform the design process of effective multimodal user interfaces.

Distributed Virtual Environments (DVEs) provide 3D graphical computer generated environments with stereo sound, supporting real-time collaboration between potentially large numbers of users distributed around the world. Early DVEs has been used over local area networks (LANs). Recently with the Internet's development into the most common embedding for DVEs these distributed applications have been moved towards an exploiting IP networks. This has brought the scalability challenges into the DVEs evolution. The network bandwidth resource is the more limited resource of the DVE system and to improve the DVE's scalability it is necessary to manage carefully this resource. To achieve the saving in the network bandwidth the different types of the network traffic that is produced by the DVEs have to be considered. DVE applications demand· exchange of the data that forms different types of traffic such as a computer data type, video and audio, and a 3D data type to keep the consistency of the application's state. The problem is that the meeting of the QoS requirements of both control and continuous media traffic already have been covered by the existing research. But QoS for transfer of the 3D information has not really been considered. The 3D DVE geometry traffic is very bursty in nature and places a high demands on the network for short intervals of time due to the quite large size of the 3D models and the DVE application requirements to transmit a 3D data as quick as possible. The main motivation in carrying out the work presented in this thesis is to find a solution to improve the scalability of the DVE applications by a consideration the QoS requirements of the 3D DVE geometrical data type. In this work we are investigating the possibility to decrease the network bandwidth utilization by the 3D DVE traffic using the level of detail (LOD) concept and the active networking approach. The background work of the thesis surveys the DVE applications and the scalability requirements of the DVE systems. It also discusses the active networks and multiresolution representation and progressive transmission of the 3D data. The new active networking approach to the transmission of the 3D geometry data within the DVE systems is proposed in this thesis. This approach enhances the currently applied peer-to-peer DVE architecture by adding to the peer-to-peer multicast neny_ork layer filtering of the 3D flows an application level filtering on the active intermediate nodes. The active router keeps the application level information about the placements of users. This information is used by active routers to prune more detailed 3D data flows (higher LODs) in the multicast tree arches that are linked to the distance DVE participants. The exploration of possible benefits of exploiting the proposed active approach through the comparison with the non-active approach is carried out using the simulation­based performance modelling approach. Complex interactions between participants in DVE application and a large number of analyzed variables indicate that flexible simulation is more appropriate than mathematical modelling. To build a test bed will not be feasible. Results from the evaluation demonstrate that the proposed active approach shows potential benefits to the improvement of the DVE's scalability but the degree of improvement depends on the users' movement pattern. Therefore, other active networking methods to support the 3D DVE geometry transmission may also be required.

This dissertation is concerned with adaptation, in the context of new emerging digital media platforms. The project proposes that new media has allowed for the creation of a universal digital heteroglossia; a heteroglossia that contains the plurality of the unstable utterances of cinema, radio, television, the web and computer games. This has allowed for the process of adaptation to become more instantaneous in the simultaneous deployment of narratives across the digital heteroglossia. Therefore, the process of adaptation is far more dialogical, with previous variants of narratives being ‘rewired’ and gaining an ‘afterlife’ through adaptation, and through the creation of new variants and versions. The Internet has allowed for adaptation to move into a participatory mode, where fanfic writers fill in ‘gaps’ left by the creators of televisual and filmic texts. Videogames, based on pre-existing or co-existing texts, mean that players can experience moments of supreme and non-permanent adaptation themselves. This thesis suggests that this participation has democratised adaptation, and has fundamentally altered the nature of ‘traditional’ adaptation. The thesis concludes that, due to a digital heteroglossia, ‘traditional’ adaptation will decline, as the process becomes more plural and instantaneous. With previous variants of narratives being summoned back into life - due to adaptation, remaking and refashioning - it is increasingly unlikely that ‘fidelity’ strategies of adaptation will continue to be the dominant discourse, as all variants of narratives begin to exist in a dialogical plurality with one another; a mutual exchange of fluctuating source and target texts, cross-referenced through intertextuality and assembling a collage of influences.

The overall aim of the research was to develop a low cost, tailorable, formative evaluation method for web designers. The last fifteen years has witnessed the rapid development of the World Wide' Web, an information resource that users surf and mine for work, leisure, entertainment and transaction purposes. Its evaluation as a computer based product has led to the assumption by some researchers, that web site and HCI usability are the same (Nielsen, 2000; Brinck, Gergle and Wood, 2002). This is reflected in many web site evaluation guidelines and instruments that characterise usability in terms of learnability, efficiency, memorability, errors and satisfaction (Rubin, 1994). However, the web is used differently from other software e.g. surfing the web is not necessarily task directed and users may access the Internet purely to obtain intellectual and emotional gratification (Spool et ai, 1999). Therefore, it is quite possible that what determines user satisfaction is very different in each context. Additionally, commerce plays a fundamental part in a lot of websites so that their goals may be different from conventional computer system design e.g. to increase the user base or encourage repeat visits. Further, with limited budgets, website designers are unlikely to involve their users during the design process and not all owebsite designers have access to an evaluator, appropriate testing facilities or evaluation knowledge to support their design. Given these arguments, this research developed and validated a low-cost evaluation 'ethod for web designers to use for formative evaluation, prior to site launch. The method addressed both HCI and commercial website goals such as the encouragement of repeat visits. To achieve this, a series of studies were conducted. Firstly, contemporary evaluation methods were evaluated to explore their possible limitations with respect to website design. T.hen the users' and designers' needs from websites and website evaluation methods were identified. This led to the development of a tai/orable method that could be applied to a wide range of small commercial websites, that took into account the issues users of websites were concerned about, and also designers' requirements of an evaluation method. The method was formalized as a set of guidelines, verified in the evaluation of a website. The potential usefulness of the method was then verified by demonstrations to selected website designers who confirmed it was useful, practical, and needed. In summary, this research contributes to knowledge firstly by identifying the need for 'designer friendly' evaluation methods in this context; secondly be showing the mismatch between current methods (based on Hel) and the requirements of website design and thirdly by producing a method to meet the need.

Augmented Reality systems, which overlay virtual information over the real world, can benefit greatly from the techniques established by the Open Hypermedia research field. Storing information and links separately from a document can be advantageous for augmented reality applications and can enable the adaption of content to suit users’ preferences. This thesis explores how Open Hypermedia systems might be used as the information systems behind AR environments. This provides benefits to augmented reality developers, not only because of the existing Open Hypermedia methods but also because of the applicability of Open Hypermedia interaction techniques to the augmented reality domain. Tangible augmented reality techniques, in which graphics are overlaid on physical objects that can be manipulated as input devices, can be used to interact with the resulting information spaces by exposing the adaptation processes in the Open Hypermedia systems. This thesis describes the development of various physical interaction metaphors that allow users to physically manipulate the underlying hypermedia structures to their liking, resulting in a natural and intuitive way to navigate complex information spaces.

Hypermedia is the vision of the complete integration of all information in any media, including text, image, audio and video. The depth and diversity of the World-Wide Web, the most successful and farthest-reaching hypermedia system to date, has tremendous potential to provide such an integrated docuverse. This thesis explores the issues and challenges surrounding the realisation of this potential through the process of Associative Writing - the authoring and publishing of integrated hypertexts which connect a writer's new contributions to the wider context of relevant existing material. Through systematically examining archived Web pages and carrying out a real-world case study, this work demonstrates that Associative Writing is an important and valid process, and furthermore that there is (albeit limited) evidence that some writers are adopting Associative Writing strategies in the Web. However, in investigating the issues facing these writers, five core challenges have been identified which may be barriers to a more widespread adoption of Associative Writing: (1) the lost in hyperspace problem, (2) legal issues over deep linking to copyrighted material, (3) the limitations of the Web hypertext model, (4) Web link integrity, and (5) that popular word-processor based Web writing tools do not adequately support each of the writing activities involved in Associative Writing. In response to these challenges, this thesis introduces the Associative Writing Framework, building on open hypertext, Semantic Web, hypertext writing, and hypertext annotation work to provide a novel interface for supporting browsing, reading, annotation, linking, and integrated writing. Although conceived in terms of supporting a generic Associative Writing scenario, the framework has been applied to the specific domain of intertextual dance analysis in order to carry out a focused evaluation. Initial indications are that the framework method is valid, and that continued work to promote and evaluate its more general applicability is worthwhile.

The amount of information on the web is characterised by being enormous, as is the number of users with different goals and interests. User models have been utilized by adaptive hypermedia systems generally and adaptive educational hypermedia systems (AEHS) particularly to personalize the amount of information they have with respect to each individual's knowledge, background and goals. As a result of the research described herein, a user model called the Hybrid Model has been developed. This model is both generic and abstract, and it extends other models used by AEHS by measuring users' knowledge levels with respect to different knowledge domains simultaneously by utilising well known techniques in the world of user modelling, specifically the Overlay model (which has been modified) and the Stereotype model. Therefore, using the Hybrid Model, AEHS will not be restricted to a single knowledge domain at anyone time. Thus, by implementing the Hybrid model, those systems can manage users' knowledge globally with respect to the deployed knowledge domains. The model has been implemented experimentally in an educational hypermedia system called WHURLE (Web-based Hierarchal Universal Reactive Learning Environment) to verify its aim - managing users' knowledge globally. Moreover, this implementation has been tested successfully through a user trial as an adaptive revision guide for a Biological Anthropology Course. Furthermore, the infrastructure of the WHURLE system has been modified to embrace the objective of the Hybrid Model. This has led to a novel design that provides the system with the capability of utilising different user models easily without affecting any of its component modules.

Information retrieval from multimedia databases has become an urgent problem. Its solution can be facilitated by describing the content of multimedia databases using a variety of ways. In a video database, the options can be caption, speech, audio, image features etc. Presently, the MPEG-7 framework deals with standardisation of the multimedia content description techniques. Image features, such as motion, colour, texture and shape, are used for image annotation. The research described here is concerned with the annotation of sports video as part of an EU project ASS AVID. The framework of ASSAVID is similar to MPEG-7. The focus of the research is to develop motion feature descriptors. Motion description becomes increasingly attractive because motion features encapsulate temporal information. However, problems plaguing low-level motion processing impede the research on high-level motion analysis. This becomes more severe in applications with real-life video. In our research, human motion is adopted for sports annotation because sports involve a number of human behaviours. Human motion analysis has a wide spectrum of applications, such as surveillance, medical imaging and information retrieval. Yet there are no techniques directly related to this topic in MPEG-7. One of the useful descriptor of complex human motion is motion periodicity. However, among the existing techniques, only a few successful attempts at periodic motion description have been reported in real-life video. In this thesis, we present a novel method for sports video retrieval using periodic motion features. We focus on modelling human motion and this is accomplished by solving several sub-problems: A novel non-rigid foreground moving object detection algorithm is developed for complex real-life video. The algorithm is used to process low- level motion and segment out the human body from images with least computational expense. Innovative sport templates are constructed for human behaviour description using periodic motion features. They represent sport types in ASSAVID. Motion feature vectors are built using the templates. Motion feature classification is accomplished using a neural network. The proposed method has been tested on the ASSAVID database, which contains more than 800 minutes of real-life video from the BBC 1992 Barcelona Olympic Games. In total about 810,000 images have been processed to test motion features. Four types of different sports are tested. The experimental results show the proposed method to be successful.

Previous research on multimedia information filtering has mainly concentrated on key frame identification and video skim generation for browsing purposes, however applications requiring the generation of summaries as the final product for user con- sumption are of equal scientific and commercial interest. Recent advances in computer vision have enabled the extraction of semantic events from an audio-visual signal, so it can be assumed for our purposes that such semantic labels are already available for use. We concentrate instead on developing methods to prioritise these semantic elements for inclusion in a summary which can be personalised to meet a particular user's needs. Our work differentiates itself from that in the literature as it is driven by the results of a knowledge elicitation study with expert summarisers. The experts in our study believe that summaries structured as a narrative are better able to convey the content of the original data to a user. Motivated by the information filtering problem, the primary contribution of this thesis is the design and implementation of a system to summarise sequences of events by automatic modelling of the causal relationships between them. We show, by com- parison against summaries generated by experts and with the introduction of a new coherence metric, that modelling the causal relationships between events increases the coherence and accuracy of summaries. We suggest that this claim is valid, not only in the domain of soccer highlights generation, in which we carry out the bulk of our experiments, but also in any other domain in which causal relationships can be iden- tified between events. This proposal is tested by applying our summarisation system to another, significantly different domain, that of business meeting summarisation, using the soccer training set and a manually generated ontology mapping. We introduce the concept of a context-group of causally related events as a first step towards modelling narrative episodes and present a comparison between a case based reasoning and a two-stage Markov model approach to summarisation. For both methods we show that by including entire context-groups in the summary, rather than single events in isolation, more accurate summaries can be generated. Our approach to personalisation biases a summary according to particular narrative plotlines using different subsets of the training data. Results show that the number of instances of certain event classes can be increased by biasing the training set appropriately. This method gives very similar results to a standard weighting method, while avoiding the need to tailor the weights to a particular application domain.

With the incorporation of free desktop videoconferencing (DVC) software on the majority of the world's PCs, over the recent years, there has, inevitably, been considerable interest in using DVC over the Internet. The growing popularity of DVC increases the need for multimedia quality assessment. However, the task of predicting the perceived multimedia quality over the Internet Protocol (IP) networks is complicated by the fact that the audio and video streams are susceptible to unique impairments due to the unpredictable nature of IP networks, different types of task scenarios, different levels of complexity, and other related factors. To date, a standard consensus to define the IP media Quality of Service (QoS) has yet to be implemented. The thesis addresses this problem by investigating a new approach to assess the quality of audio, video, and audiovisual overall as perceived in low cost DVC systems. The main aim of the thesis is to investigate current methods used to assess the perceived IP media quality, and then propose a model which will predict the quality of audiovisual experience from prevailing network parameters. This thesis investigates the effects of various traffic conditions, such as, packet loss, jitter, and delay and other factors that may influence end user acceptance, when low cost DVC is used over the Internet. It also investigates the interaction effects between the audio and video media, and the issues involving the lip sychronisation error. The thesis provides the empirical evidence that the subjective mean opinion score (MOS) of the perceived multimedia quality is unaffected by lip synchronisation error in low cost DVC systems. The data-gathering approach that is advocated in this thesis involves both field and laboratory trials to enable the comparisons of results between classroom-based experiments and real-world environments to be made, and to provide actual real-world confirmation of the bench tests. The subjective test method was employed since it has been proven to be more robust and suitable for the research studies, as compared to objective testing techniques. The MOS results, and the number of observations obtained, have enabled a set of criteria to be established that can be used to determine the acceptable QoS for given network conditions and task scenarios. Based upon these comprehensive findings, the final contribution of the thesis is the proposal of a new adaptive architecture method that is intended to enable the performance of IP based DVC of a particular session to be predicted for a given network condition.

Open hypermedia concerns itself with finding ways to provide links between information using systems that allow the direct manipulation of links. These systems are generally built in component fashion, a major component will be the Link Service. This thesis extends previous work on building link services for the World Wide Web in order to improve the quality of linking that can be provided. A link service delivers links into Web pages dynamically, this allows the system the opportunity to adapt the links to suit the user and content. The more general form of this adaption is to deliver links in some context. The core of this work is to examine how to generalise the delivery of links in context so that new context models and techniques can be introduced without writing a new link service each time. This thesis examines the problem of defining models of context and how different models are often incompatible. It introduces a new architecture for exploring methods of providing links in context. The architecture consists of a core infrastructure along with a method for dynamically changing the components of the system concerned with the link model and the computation process for generating the final link to deliver. The effectiveness of the architecture is explored through the undertaking of a number of projects using real-world data. The limits of the design are explored and the analysis focuses on how successful it is possible to be given the goals set.

The consequences of adapting the learning style and learning strategies differences have not been thoroughly pursued by adaptive hypermedia researchers. The undesired effect of 'ordinary' hypermedia learning systems is that they may not effectively target the students' learning styles. Many researchers recognise the importance of personalising hypermedia to meet students' needs and the literature review suggests that students perform better in hypermedia based learning environments that are configured to support their learning needs. Two novel adaptive systems have been developed, based on the theory of learning styles and learning strategies. This first part of the investigation presents the findings of an evaluation of a hypermedia learning system that incorporated global and sequential learning styles. Details of the design regarding the structure of the information and linking mechanisms are presented. A review of the results obtained by evaluating the use of the system suggests that students, who use hypermedia-learning environments that supported their preferred learning style, performed significantly better. Building on this learning styles research, the study presents the concept and design involved in building an adaptive hypermedia system that supports application and use of "higher order" learning strategies, such as summarising and questioning. The system has been named ILASH, standing for Incorporating Learning Strategies in Hypermedia. The thesis presents the findings of the evaluation of the system. Students are expected to achieve significant improvement in learning outcomes while interacting with the adaptive version of the system compared to the non-adaptive version. The results obtained from the summative evaluation suggest that the adaptive features provided by the system have significantly contributed towards improved learning performance. The development of the system emphasises the fact that adaptive educational systems can match some of the students' learning preferences and may have implications on the design of web based educational courseware.

This thesis considers the potential of interactive multimedia as a new medium for communicating ideas and arguments. In focusing on the relationship between interactive multimedia and anthropology, the aim is to look at possibilities for the creation of single authored texts which represent aspects of long-term studies of cultural change. Using data gained from empirically based fieldwork study, the thesis considers how photographic and moving image records of culture might be presented alongside descriptive and analytical text to communicate ideas and arguments. Key issues raised by the thesis relate to the problems and possibilities involved in using interactive multimedia to seamlessly integrate diverse forms media, and in giving the user a sense of agency without compromising the integrity of the author's ideas and arguments. Possibilities for creating both tightly structured and more open-ended forms are considered. A key possibility that is explored relates to ways in which authors can use the medium to communicate discrete points by creating sets of moving image juxtapositions which can be linked together to create an argument. It is argued that this enables ideas and arguments to be presented using techniques of comparison, offering an alternative to the moving image convention of sequential editing. I suggest that this creates new possibilities for scholarly discourse which are not necessarily predicated on the concept of a linear Chain of reasoning, and I relate these possibilities to well-established debates about the use of moving images within anthropology. In so doing, I suggest new ways in which moving image records might be used to help deliver anthropological ideas and arguments as opposed to serving as an illustrative adjunct to text-based discourse. The thesis has been written from an interdisciplinary perspective, combining debates within anthropology with issues relating to media production.

The purpose of this research was to examine the control of virtual environment (VE) navigation and interaction tasks for people with moderate to severe learning difficulties. This research stems from the development of VEs for people with learning difficulties and the identification that there are usability difficulties with the computer input devices, which are used to control the VE tasks. This investigation resulted in the following main extensions to the field of study: • The application of a multi-disciplinary design methodology, resulting in the development of a design specification for the selection or design of usable VE input devices for young people with moderate to severe learning difficulties. • The design and development of a new VE input system for young people with moderate to severe learning difficulties. An evaluation was carried out to identify the specific usability difficulties that young people with moderate to severe learning difficulties experience, when using the joystick and mouse (commonly used devices) to control VE tasks. This evaluation concluded that it is important to consider user abilities, the VE tasks and the working environment, in order to select or develop usable VE input devices. This conclusion was supported by the following disciplines: humancomputer interaction (HCI); user-centred design (UCD) and assistive technology (AT). Consequently, a multi-disciplinary design methodology, which was based on the key activities of the user-centred design (UCD) process, was carried out. The first step of this methodology was 'understand and specify the context of use', which involved the analysis of the users, the tasks and the working environment, hence satisfying the conclusion of the background research. A new VE input system (VRl) was developed, based on the user, task and environment research. This prototype was then tested with a user group of young people with moderate to severe learning difficulties to evaluate its usability and to test the following hypothesis: • The employment of the multi-disciplinary design methodology results in the design and development of a VE input system for young people with moderate to severe learning disabilities, which has greater usability than a commonly used system for this user population. In order to test this hypothesis, a comparison was made between the usability of VRl and the JM system (Joystick and Mouse: commonly used devices). The results showed VRl to be significantly more usable than the JM system. Hence, the hypothesis was supported. The results of this research have shown that the multi-disciplinary design methodology employed was a successful approach to this investigation. The application of this methodology has resulted in the production of a VE input system (VRl) that has the potential to provide improved VE control for young people with moderate to severe learning difficulties. Due to time limitations it was not possible to incorporate all of the identified design requirements into the VRl prototype. Hence, to further increase usability, future prototype development should apply the complete list of device attributes from the design specification, along with the design refinement identified from the user-based assessment.