PhD Theses

Swati Agarwal, Open source social media as sensors for enabling government identification, prediction and response applications, Indraprastha Institute of Information Technology, Delhi (IIIT-D), India, Jun 2017, PhD.
Abstract | Full document

Online Social media platforms such as Tumblr, Twitter (micro-blogging website) and YouTube (video sharing website) contains information which is publicly available or open-source. Open source social media intelligence (OSSMInt) is a field comprising of techniques and applications to analyze and mine open-source social media data for extracting actionable information and useful insights. The focus of the work presented in this dissertation is on novel applications and techniques of OSSMInt in the government sector. We propose and develop several novel usage scenarios and applications around OSSMInt for government and broadly divide them into three categories: identification, prediction, and response applications. In particular, we present solutions, tools and techniques for analyzing data from micro-blogging website to analyze citizen complaints and grievances in the public sector [response]. The research presented in this dissertation also describes our work on analyzing data from Twitter micro-blogging website to early forecast a civil unrest and protest [prediction]. Furthermore, we build various applications around identification and detection that are useful for the government and security analysts. We demonstrate the application of OSSMInt for identifying religious conflicts within society by mining public opinions on Tumblr website and fill the gaps of offline surveys. The study presented in this dissertation propose solutions for enabling law enforcement agencies to detect, prevent and combat online radicalization and extremism (content, users, and communities) by mining data from Tumblr, Twitter and YouTube [identification]. We also propose to use the deep natural language processing analysis based techniques for automatic identification of racist and radicalized posts based on the intent of the author. Furthermore, we also propose and build an application for detecting secret message exchanged in an adversarial communication and capture the obfuscated terms in messages. It is technically challenging to analyze social media content due to the free-form nature of user-generated data that raises several issues such as incorrect grammar, spelling mistakes, multilingual scripts, term obfuscation and usage of abbreviation and short-forms. In this dissertation, we present several techniques for data processing, text classification, and word obfuscation detection and information extraction for overcoming the noisy data problem. We also propose computational linguistic-based methods to address the challenges of ambiguity in the textual content. The central component of our proposed solution approach is the application of information retrieval and machine learning based techniques and algorithms. Our study consists of experimenting with a diverse range of machine learning algorithms such as unsupervised, semi-supervised and supervised learning (k-NN, SVM, Naive Bayes, Random Forest and Decision Tree) based algorithms. We also employ several ensemble learning based technique to improve the accuracy and performance of the baseline statistical models. We make the processed dataset used in our experiments publicly available for other researchers to replicate our experiments and benchmark against our proposed techniques. Data visualization is one of the major components of data analysis and interpretation. The study employs several basic and advanced data visualization techniques to present information in an intuitive manner to the end user.

Rodrigo Laiola Guimarães, Socially-Aware Multimedia Authoring, Vrije Universiteit Amsterdam, Jan 2014, PhD.
Abstract | Full document

Creating compelling multimedia productions is a non-trivial problem. This is true for both professional and personal content. For professional content, extensive production support is typically available during creation. Content assets are well structured, content fragments are professionally produced with high quality, and production assets are often highly annotated (within the scope of the production model). For personal content, nearly none of these conditions exist: content is a collection of assets that are structured only by linear recording time, of mediocre technical quality (on an absolute scale), and with only basic automatic annotations. These conditions limit the options open to casual authors and to viewers of rich multimedia content in creating and receiving focused, highly personal media presentations. The problem is compounded when authors want to integrate community media assets: media fragments donated from a potentially wide and anonymous recording community. In this thesis we reflect on the traditional multimedia authoring workflow and we argue that a fresh new look is required. Our experimental methodology aims at meeting the requirements needed for social communities that are not addressed by traditional authoring and sharing applications. We focus on the particular task of supporting socially-aware multimedia authoring, in which the relationships within particular social groups can be exploited to create highly personal media experiences. Our framework is centered on empowering users in telling stories and commenting on personal media artifacts, considering the long-term social context of the user. The work has been evaluated through a number of prototype tools that allow users to explore, create, enrich and share rich multimedia artifacts. Results from our evaluation process provide useful insights into how a socially-aware multimedia authoring and sharing system should be designed and architected, for helping users in recalling personal memories and in nurturing their close circle relationships.

Olaf Hartig, Querying a Web of Linked Data: Foundations and Query Execution, Humboldt-Universität zu Berlin, Aug 2014, PhD.
Abstract | Full document

During recent years a set of best practices for publishing and connecting structured data on the World Wide Web (WWW) has emerged. These best practices are referred to as the Linked Data principles and the resulting form of Web data is called Linked Data. The increasing adoption of these principles has lead to the creation of a globally distributed space of Linked Data that covers various domains such as government, libraries, life sciences, and media. Approaches that conceive this data space as a huge distributed database and enable an execution of declarative queries over this database hold an enormous potential; they allow users to benefit from a virtually unbounded set of up-to-date data. As a consequence, several research groups have started to study such approaches. However, the main focus of existing work is to address practical challenges that arise in this context. Research on the foundations of such approaches is largely missing. This dissertation closes this gap. This dissertation first establishes a well-defined framework for defining and studying queries over Linked Data on the WWW. In particular, we introduce a data model that enables us to formally conceive Linked Data on the WWW as a (distributed) database and a computation model that captures the capabilities of a query execution system for this database. Based on these models, we adapt the declarative query language SPARQL to the given scenario. More precisely, we define a full-Web query semantics and a family of reachability-based query semantics such that each of these query semantics presents a well-defined basis for using SPARQL to query Linked Data on the WWW. Thereafter, we show theoretical properties of queries under these query semantics. Perhaps the most important result of this study formally verifies the common assumption that a computation of query results that are complete w.r.t. all Linked Data on the WWW, is not feasible. However, we also identify classes of queries for which the computational feasibility is less limited. After analyzing queries over Linked Data independent of specific approaches for executing such queries, this dissertation focuses on a particular execution approach and studies fundamental aspects thereof. The studied approach presents a general strategy for executing queries by integrating traversal-based data retrieval into the result construction process. To analyze this notion of traversal-based query execution formally, we define it in the form of an abstract query execution model. In addition, we discuss a concrete implementation approach for our execution model; this approach is based on the well-known concept of iterators. Our analysis of both the execution model and the iterator-based implementation shows that (i) for one of our reachability-based query semantics, the given notion of traversal-based query execution, in general, is sound and complete, whereas (ii) for the same query semantics, the specific, iterator-based implementation approach cannot guarantee completeness of query results. Based on an experimental analysis we verify that the latter limitation has a significant impact in practice.

Jérôme Kunegis, On the Spectral Evolution of Large Networks, University of Koblenz–Landau, Nov 2011, PhD.
Abstract | Full document

In this thesis, I study the spectral characteristics of large dynamic networks and formulate the spectral evolution model. The spectral evolution model applies to networks that evolve over time, and describes their spectral decompositions such as the eigenvalue and singular value decomposition. The spectral evolution model states that over time, the eigenvalues of a network change while its eigenvectors stay approximately constant. I validate the spectral evolution model empirically on over a hundred network datasets, and theoretically by showing that it generalizes a certain number of known link prediction functions, including graph kernels, path counting methods, rank reduction and triangle closing. The collection of datasets I use contains 118 distinct network datasets. One dataset, the signed social network of the Slashdot Zoo, was specifically extracted during work on this thesis. I also show that the spectral evolution model can be understood as a generalization of the preferential attachment model, if we consider growth in latent dimensions of a network individually. As applications of the spectral evolution model, I introduce two new link prediction algorithms that can be used for recommender systems, search engines, collaborative filtering, rating prediction, link sign prediction and more. The first link prediction algorithm reduces to a one-dimensional curve fitting problem from which a spectral transformation is learned. The second method uses extrapolation of eigenvalues to predict future eigenvalues. As special cases, I show that the spectral evolution model applies to directed, undirected, weighted, unweighted, signed and bipartite networks. For signed graphs, I introduce new applications of the Laplacian matrix for graph drawing, spectral clustering, and describe new Laplacian graph kernels. I also define the algebraic conflict, a measure of the conflict present in a signed graph based on the signed graph Laplacian. I describe the problem of link sign prediction spectrally, and introduce the signed resistance distance. For bipartite and directed graphs, I introduce the hyperbolic sine and odd Neumann kernels, which generalize the exponential and Neumann kernels for undirected unipartite graphs. I show that the problem of directed and bipartite link prediction are related by the fact that both can be solved by considering spectral evolution in the singular value decomposition.

Domenico Gendarmi, Collaborative Tagging as a Community-Driven Approach to Knowledge Sharing, University of Bari, Apr 2010, PhD.
Abstract

Knowledge is strongly tied up with cognitive and social aspects, as the management of knowledge occurs within a tangled structured social context. Human and social factors involved in the exchange of knowledge have thus a strong influence on effectiveness of any knowledge sharing process. One of the biggest challenges for supporting knowledge sharing in online communities is still fostering the individual‚s willingness to share knowledge with other members. Although there have been a number of approaches to support knowledge sharing in communities, they all have been affected by a significant disparity of effort versus benefit from an individual perspective. Moreover, some traditional approaches at the base of existing knowledge management systems show elitist attitudes, which usually hinder a large adoption of such authoritative knowledge bases for sharing information within Web communities. This thesis embraces the idea that an opportunity to overcome some hindrances to knowledge sharing in online communities is to involve users directly in the knowledge management life cycle, since its creation, until the sharing stage. Nevertheless, in order to encourage an active participation by members of an online community to this process, people need ways to find required information easily and to perceive personal benefits from the sharing of retrieved information. In the last years the phenomenon of Social Web has arisen as a class of Web applications in which one of the most distinctive features is the active users‚ participation. Social Web focuses on creating knowledge through collaboration and the social interactions of individuals. A prominent example of Social Web applications are collaborative tagging systems, also known as folksonomies. Collaborative tagging systems let individuals organize information they found on the Web using terms that reflect personal assertions about resources. The research purposes of this thesis focus on supporting individual contributions in a dynamic knowledge-sharing environment that relies on user-generated tags as a simple way to privately organize information and connect users with similar interests. In order to encourage and exploit the active participation of people in the collective process of knowledge sharing, this thesis proposes and evaluates two different tools. On one hand, a tool that aggregates users‚ contributions coming from different communities of readers and publishes such content as re-usable linked data on the Web. On the other hand, this thesis proposes a tag recommender system which relies on the semantic content analysis of the information to be organized as well as on the personal and collective tagging history within a social bookmarking system.

Shamim Bodhanya, Strategic Enactment: An Interpretive Approach To Organisational Strategy, University of KwaZulu-Natal, Apr 2010, PhD.
Abstract

The purpose of this thesis is to explore the field of strategy by way of its historical trajectory and to consider the major branches that constitute this broad, but fragmented discipline. It is an interdisciplinary endeavour that draws specifically on systems theories and complexity theory as a way to enrich the field. The strategy field tends to be philosophically unreflexive. As a result it is dominated by an objectivist ontology, which underpins strategic choice. One of the aims of this thesis is to explore the implications for strategy, if instead, an interpretive stance, based on an ontology of social constructionism, is adopted. The literature has not fully explored and developed different ontologies in the context of strategy and hence has left a major gap in theorising about strategy. This thesis attempts to address that gap and therefore one of the contributions of the study will be a tentative theory of strategic enactment. This research attempts to answer the following key questions: 1. What are the major theoretical frameworks and conceptual models that frame the field of strategy? 2. How well do these frameworks and models contribute to strategy under conditions of high ambiguity and uncertainty? 3. What contributions may be made by applying complexity theory to the field of strategy? 4. What are the implications of adopting an interpretive approach to strategy? 5. What are the implications of strategic enactment on strategic leadership? Given that these research questions are of a philosophical and theoretical nature, the research methodology and approach is one based on theoretical exploration. It is therefore not an empirical study, but a conceptual one embracing both breadth and depth. It is broad in that it covers multiple literature sets which include bodies of knowledge in organisational theory, leadership, strategy, systems thinking and complexity theory. It is deep in its interrogation of core conceptual constructs that are pertinent to the strategy frame of reference and in its comprehensive coverage of the major topics that circumscribe the field. While it relies on an extensive coverage of existing texts it is not a hermeneutic study from a methodological point of view. It does not purport to interpret and to elicit the meaning of texts. The term interpretive in the title instead refers to the ontological notion of sensemaking and interpretation that is central to strategic enactment. Interpretive in this sense is not an interpretation of texts in a hermeneutic fashion, but interpretive in relation to enacting reality. Despite being a theoretical study it still draws on deductive, inductive and abductive reasoning.

Alvin Chin, Social Cohesion Analysis of Networks: A Novel Method for Identifying Cohesive Subgroups in Social Hypertext, University of Toronto, Jan 2009, PhD.
Abstract

Finding subgroups within social networks is important for understanding and possibly influencing the formation and evolution of online communities. This thesis addresses the problem of finding cohesive subgroups within social networks inferred from online interactions. The dissertation begins with a review of relevant literature and identifies existing methods for finding cohesive subgroups. This is followed by the introduction of the SCAN method for identifying subgroups in online interaction. The SCAN (Social Cohesion Analysis of Networks) methodology involves three steps: selecting the possible members (Select), collecting those members into possible subgroups (Collect) and choosing the cohesive subgroups over time (Choose). Social network analysis, clustering and partitioning, and similarity measurement are then used to implement each of the steps. Two further case studies are presented, one involving the TorCamp Google group and the other involving YouTube vaccination videos, to demonstrate how the methodology works in practice. Behavioural measures of Sense of Community and the Social Network Questionnaire are correlated with the SCAN method to demonstrate that the SCAN approach can find meaningful subgroups. Additional empirical findings are reported. Betweenness centrality appears to be a useful filter for screening potential subgroup members, and members of cohesive subgroups have stronger community membership and influence than others. Subgroups identified using weighted average hierarchical clustering are consistent with the subgroups identified using the more computationally expensive k-plex analysis. The value of similarity measurement in assessing subgroup cohesion over time is demonstrated, and possible problems with the use of Q modularity to identify cohesive subgroups are noted. Applications of this research to marketing, expertise location, and information search are also discussed.

Benjamin Markines, Socially Induced Semantic Networks and Applications, Indiana University, May 2009, PhD.
Abstract | Full document

Social bookmarking systems allow Web users to actively annotate online resources. These annotations incorporate meta-information with Web pages in addition to the actual document contents. From a collection of socially annotated resources, we present various methods for quantifying the relationship between objects, i.e., tags or resources. These relationships can then be represented in a semantic similarity network where the nodes represent objects and the undirected weighted edges represent their relations. These relations are quantified through similarity measures. There are two challenges associated with assembling and maintaining such a similarity network. The first challenge is updating the relations efficiently, i.e., the time and space complexity associated with graph algorithms. The complexity of these algorithms is typically quadratic. We present an incremental process answering both space and time limitations. The second challenge is the quality of the similarity measure. We evaluate various measures through the approximation of reference similarities. We then present a number of applications leveraging socially induced semantic similarity networks. A tag recommendation system, a page recommendation engine, and a Web navigation tool are evaluated through user studies. Finally, we design spam detection algorithms to enhance the functionality of social bookmarking systems.

Ching-man Au Yeung, From User Behaviours to Collective Semantics, University of Southampton, Oct 2009, PhD.
Abstract

The World Wide Web has developed into an important platform for social interactions with the rise of social networking applications of different kinds. Collaborative tagging systems, as prominent examples of these applications, allow users to share their resources and to interact with each other. By assigning tags to resources on the Web in a collaborative manner, users contribute to the emergence of complex networks now commonly known as folksonomies, in which users, documents and tags are interconnected with each other. To reveal the implicit semantics of entities involved in a folksonomy, one requires an understanding of the characteristics of the collective behaviours that create these interconnections. This thesis studies how user behaviours in collaborative tagging systems can be analysed to acquire a better understanding of the collective semantics of entities in folksonomies. We approach this problem from three different but closely related perspectives. Firstly, we study how tags are used by users and how their different intended meanings can be identified. Secondly, we develop a method for assessing the expertise of users and quality of documents in folksonomies by introducing the notion of implicit endorsement. Finally, we study the relations between documents induced from collaborative tagging and compare them with existing hyperlinks between Web documents. We show that, in each of these scenarios, it is crucial to consider the collective behaviours of the users and the social contexts in order to understand the characteristics of the entities. This project can be considered as a case study of the Social Web, the research outcomes of which can be easily generalised to many other social networking applications. It also fits into the larger framework for understanding the Web set out by the emerging interdisciplinary field of Web Science, as the work involves analyses of the interactions and behaviour of Web users in order to understand how we can improve existing systems and facilitate information sha ring and retrieval on the Web.

Markel Vigo, Automatic Assessment of Contextual Web Accessibility from an Evaluation, Measurement and Adaptation Perspective, University of the Basque Country, Nov 2009, PhD.
Abstract | Full document

Web accessibility aims at providing people with disabilities with a barrier-free user experience. This PhD report explores how automatic accessibility assessment benefits both accessibility experts and end-users. In the context of this document assessment is understood as a twofold goal consisting of evaluation and measurement. The former group, accessibility experts, can make use of a set of tools and flexible frameworks to create and maintain accessible content for a wide range of users and interaction environments. In addition, the techniques and methods used throughout this document can be replicated and incorporated by third parties. Expert assessment of fourteen websites and large- scale automatic assessment of almost 1,500 web pages demonstrates that the Web Accessibility Quantitative Metric (WAQM) is a valid and reliable metric to automatically measure accessibility of web pages. Moreover, by deploying WAQM jointly with commercial search engines (Google and Yahoo!) it is corroborated that search engine crawlers consider web accessibility while indexing the Web although pages are not ranked according to their accessibility score. The latter group, end-users, can take advantage from contextual accessibility assessment by deploying assessment scores in web pages to increase user orientation. To do so from a guidelines conformance point of view, not only content accessibility has to be considered but also the dependencies between content guidelines, user group, assistive technology and the access device. Twenty users took part in a user test conducted for determining the validity of this approach with the mobile web, concluding that device-tailored assessment is more faithful to usability assessment in terms of satisfaction and task completion time. Another experiment with sixteen blind users also demonstrates that deploying user-tailored scores increases user orientation and confidence. It is also observed that when links are annotated with the accessibility score of the page they point to, blind users change the browsing paradigm from sequential navigation to random criteria within those links that score higher. Finally it is noteworthy that lessons drawn from automatic assessment can be extrapolated to manual assessment carried out by experts or user-testing.

Thanyalak Maneewatthana, Contextual Annotation Framework using Ontology-Based Contextual Annotation Service, University of Southampton, Apr 2008, PhD.
Abstract | Full document

Current systems do not provide a context when sharing and reusing annotations. This research investigated how this may be achieved. This thesis provides an overview of the work that has been undertaken in the fields of Semantic Web, adaptive hypermedia system, and open hypermedia. This work discusses experiments that were conducted to assess the benefit of integrating Semantic Web technologies with open hypermedia and adaptive hypermedia concepts to provide adaptable hypermedia for usersÂ’ requirements. In this work, ontologies are used to explicitly define models, concepts, user profiles, context and semantic relationships. To enable sharing and reusability of information chunks as annotations, this research brings together several technologies: ontologies for knowledge representation, and extended FOHM to represent the structure of annotations. A contextual annotation framework (CAF) using an ontology-based contextual annotation service is proposed. The novel contribution of this work is introducing the CAF using an ontology-based contextual annotation service by building on open hypertext and the Semantic Web. The a-PIE is prototyped to provide a system for supporting browsing, reading, annotating hypertext, and manipulating interesting annotations in a personal repository. The framework has been applied to the specific domain of web development, in order to carry out a focused evaluation. The results indicate that the framework is valid.

T. Chithralekha, Agents with Two-Dimensional Language Autonomy for Task Delegation, Pondicherry University, Aug 2008, PhD.
Abstract

A software agent acts on behalf of someone to carry out a particular task which has been delegated to it. In order to fulfill the interaction requirements of delegation, the agent needs to provide for collaborative natural language interaction added with the feature of multilingualism to facilitate a flexible form of interaction in the language required by the user. In the existing agents, the contributions from the fields of computational linguistics and multilingual computing are rendered to provide natural language interaction and multilingualism. But, the language ability requirements of an agent, considered from an agent perspective, necessitates a revamping in the requirements from both conceptual and architectural perspectives. The process of fulfillment of requirements inevitably leads to attributing two dimensions of language autonomy namely, language ability management autonomy and language behavior autonomy to the language ability of an agent. This results in an evolved form of language ability which is comparable to that of humans and hence termed as a language faculty. The two dimensions of autonomy and the solutions found for other relevant issues are used in building the conceptual model of the language faculty. A new Belief, Task, Behavior paradigm for representing the internal states of the language faculty is derived from the conceptual model. Based on this paradigm, a new Behavior Management Architecture for the language faculty of an agent is designed. The architecture is supported with appropriate Role-Based design and Aspect-Oriented implementation models so that an agent with the required form of language ability could be realized. Using the architecture and its design and implementation models, a Multilingual Natural language Agent Interface (MULLAI) for Email server is developed to illustrate the use of the solutions developed. The developed architecture is also extended in order to attribute a language faculty suitable for a multi-agent system, namely the Multilingual Multi-Agent System (MMAS). An evaluation of the agent with two-dimensional language autonomy with the existing multilingual dialoguing agents is also performed. A secondary contribution of this work is that the developed solutions / models are generic in nature and not specific to a language faculty.

Sabrina Geissler, ‘Medial distillation’ as an innovative quality of software-based media, University of Paderborn, Jul 2008, PhD.
Abstract | Full document

The present dissertation contains an interdisciplinary approach to study aspects of digital media both from a technical as from a cultural point of view. The concept of medial distillation constitutes the core of the project: Based on the idea of the shared workspace, it offers an adequate perspective to determine innovative qualities of software-based media and to distinguish a variety of technical functions (e. g. media functions, active typography, slow and fast interaction). Therefore, the concept permits to reveal innovative qualities which are inherent to software-based media: As such media are able to condense information, they can handle quantities in a different way than traditional media do.

Angelo Di Iorio, Pattern-based Segmentation of Digital Documents: Model and Implementation, University of Bologna, Apr 2007, PhD.
Abstract | Full document

This thesis proposes a new document model, according to which any document can be segmented in some independent components and transformed in a pattern-based projection, that only uses a very small set of objects and composition rules. The point is that such a normalized document expresses the same fundamental information of the original one, in a simple, clear and unambiguous way. The central part of my work consists of discussing that model, investigating how a digital document can be segmented, and how a segmented version can be used to implement advanced tools of conversion. I present seven patterns which are versatile enough to capture the most relevant documents\' structures, and whose minimality and rigour make that implementation possible. The abstract model is then instantiated into an actual markup language, called IML. IML is a general and extensible language, which basically adopts an XHTML syntax, able to capture the only content of a digital document. It is compared with other languages and proposals, in order to clarify its role and objectives. Finally, I present some systems built upon these ideas. These applications are evaluated in terms of users\' advantages, workflow improvements and impact over the overall quality of the output. In particular, they cover heterogeneous content management processes: from web editing to collaboration (IsaWiki and WikiFactory), from e-learning (IsaLearning) to professional printing (IsaPress).

César Moura, MDEduc: Conceiving and Implementing a Language-Oriented Approach for the Design of Automated Learning Scenarios, University of Lille, Jul 2007, PhD.
Abstract | Full document

Aiming at developing learning applications, the e-learning communities have recently made a profound switch in their modus operandi. Instead of following prescriptive processes leading to turn-key applications reflecting an envisaged learning scenario, as in traditional ISD processes, learning designers now make use of Educational Modelling Languages (EML), provided by standard setting bodies along with the required tooling, so they can build their own applications. Being at close range to practitioners, learning designers will be able to build tailored applications in a fraction of the time required by traditional ISD and in a much more flexible approach - since no programming is needed. However this flexibility is limited. EMLs present a fixed vocabulary, permitting its elements to be arranged to create an instance of a scenario, but disallowing the addition of new concepts - in case existing ones do not satisfy. Since EMLs, as epitomised by the IMS-LD specification, intend to capture all aspects of educational phenomena, they tend to become either too complex or too inexpressive - in the first case becoming unwieldy for tackling any particular scenario and in the second, failing to conveniently express any reasonably rich scenario. Furthermore, it is claimed on this thesis that any particular EML cannot represent but a single aspect of learning scenarios, which in real life are multi-dimensional and always open to diverse interpretations by different people. So, this thesis sustains that the single EML approach is fundamentally flawed, and proposes an alternative one, called multi-EML, which empowers learning designers with the possibility of formulating the very conceptualisation of their problems, by allowing them to create their own EML reflecting their specific requirements - as opposed to just rearrange the concepts of a fixed language. In sum, the Multi-EML approach enables: Capturing the knowledge of domain experts in informal notations (particularly using design patterns for education, or pedagogical patterns); Starting from these original descriptions, to build formal models (or EMLS) representing their specific requirements; Generating full-fledged applications driven by the formal models.

Hend S. Al-Khalifa, Automatic Document-Level Semantic Metadata Annotation Using Folksonomies and Domain Ontologies, Southampton Univeristy, Jul 2007, PhD.
Abstract | Full document

The last few years have witnessed a fast growth of the concept of Social Software. Be it video sharing such as YouTube, photo sharing such as Flickr, community building such as MySpace, or social bookmarking such as del.icio.us. These websites contain valuable user-generated metadata called folksonomies. Folksonomies are ad hoc, light-weight knowledge representation artefacts to describe web resources using people's own vocabulary. The cheap metadata contained in such websites presents potential opportunities for us (researchers) to benefit from. This thesis presents a novel tool that uses folksonomies to automatically generate metadata with educational semantics in an attempt to provide semantic annotations to bookmarked web resources, and to help in making the vision of the Semantic Web a reality. The tool comprises two components: the tags normalisation process and the semantic annotation process. The tool uses the del.icio.us social bookmarking service as a source for folksonomy tags. The tool was applied to a case study consisting of a framework for evaluating the usefulness of the generated semantic metadata within the context of a particular eLearning application. This implementation of the tool was evaluated over three dimensions: the quality, the searchability and the representativeness of the generated semantic metadata. The results show that folksonomy tags were acceptable for creating semantic metadata. Moreover, folksonomy tags showed the power of aggregating people's intelligence. The novel contribution of this work is the design of a tool that utilises folksonomy tags to automatically generate metadata with fine gained and extra educational semantics.

J. Michael Moore, Design Exploration: Engaging a Larger User Population, A&M University, Aug 2007, PhD.
Abstract

Software designers must understand the domain, work practices, and user expectations before determining requirements or generating initial design mock-ups. Users and other stakeholders are a valuable source of information leading to that understanding. Much work has focused on design approaches that include users in the software development process. These approaches vary from surveys and questionnaires that garner responses from a population of potential users to participatory design processes where representative users are included in the design/development team. The Design Exploration approach retains the remote and asynchronous communication of surveys while making expression of feedback easier by providing users alternatives to textual communication for their suggestions and tacit understanding of the domain. To do this, visual and textual modes of expression are combined to facilitate communication from users to designers while allowing a broad user audience to contribute to software design. One challenge to such an approach is how software designers make use of the potentially overwhelming combination of text, graphics, and other content. The Design Exploration process provides users and other stakeholders the Design Exploration Builder, a construction kit where they create annotated partial designs. The Design Exploration Analyzer is an exploration tool that allows software designers to consolidate and explore partial designs. The Analyzer looks for patterns based on textual analysis of annotations and spatial analysis of graphical designs, to help identify interesting examples and patterns within the collection. Then software designers can use this tool to search and browse within the exploration set in order to better understand the task domain, user expectations and work practices. Evaluation of the tools has shown that users will often work to overcome expression constraints to convey information. Moreover, the mode of expression influences the types of information garnered. Furthermore, including more users results in greater coverage of the information gathered. These results provide evidence that Design Exploration is an approach that collects software and domain information from a large group of users that lies somewhere between questionnaires and face to face methods.

Nick Montfort, Generating Narrative Variation in Interactive Fiction, University of Pennsylvania, Aug 2007, PhD.
Abstract | Full document

A general method for the generation of natural language narrative is described. It allows the expression, or narrative discourse, to vary independently of the underlying events and existents that are the narrative's content. Specifically, this variation is accomplished in an interactive fiction (IF) system which replies to typed input by narrating what has happened in a simulated world. IF works have existed for about 30 years as forms of text-based computer simulation, instances of dialog systems, and examples of literary art. Theorists of narrative have carefully distinguished between the level of underlying content (corresponding to the simulated world in interactive fiction) and that of expression (corresponding to the textual exchange between computer and user) since the mid-1960s, when the field of narratology began to develop, but IF systems have not yet made use of this distinction. The current project contributes new techniques for automatic narration by building ! on work done in computational linguistics, specifically natural language generation, and in narratology. First, types of narrative variation that are possible in IF are identified and formalized in a way that is suitable for a natural language generation system. An architecture for an IF system is then described and implemented; the result allows multiple works of interactive fiction to be realized and, using a general plan for narrating, allows them to be narrated in different ways during interaction. The system's ability to generate text is considered in a pilot evaluation. Plans for future work are also discussed. They include publicly released systems for IF development and narratology education, adding a planning capability that uses actors' individual perspectives, and adapting automatic narration to different sorts of interactive systems.

Panchit Longpradit, Links Personalisation with Multi-Dimensional Linkbases, University of Southampton, Jun 2007, PhD.
Abstract | Full document

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.

Beat Signer, Fundamental Concepts for Interactive Paper and Cross-Media Information Spaces, ETH Zurich, Jan 2006, PhD.
Abstract | Full document

While there have been dramatic increases in the use of digital technologies for information storage, processing and delivery over the last twenty years, the affordances of paper have ensured its retention as a key information medium. Despite predictions of the paperless office, paper is ever more present in our daily work as reflected by the continuously increasing worldwide paper consumption. Many researchers have argued for the retention of paper as an information resource and its integration into cross-media environments as opposed to its replacement. This has resulted in a wide variety of projects and technological developments for digitally augmented paper documents over the past decade. However, the majority of the realised projects focus on technical advances in terms of hardware but pay less attention to the very fundamental information integration and cross-media information management issues. Our information-centric approach for a tight integration of paper and digital information is based on extending an object-oriented database management system with functionality for cross-media information management. The resulting iServer platform introduces fundamental link concepts at an abstract level. The iServer\'s core link management functionality is available across different multimedia resources. Only the media-specific portion of these general concepts, for example the specification of a link\'s source anchor, has to be implemented in the form of a plug-in to support new resource types. This resource plug-in mechanism results in a flexible and extensible system where new types of digital as well as physical resources can easily be integrated and, more importantly, cross-linked to the growing set of supported multimedia resources. In addition to the associative linking of information, our solution allows for the integration of semantic metadata and supports multiple classification of information units. iServer can, not only link between various static information entities, but also link to active content and this has proven to be very effective in enabling more complex interaction design. As part of the European project Paper++, under the Disappearing Computer Programme, an iServer plug-in for interactive paper has been implemented to fully integrate paper and digital media, thereby gaining the best of the physical and the digital worlds. It not only supports linking from physical paper to digital information, but also enables links from digital content to physical paper or even paper to paper links. This multi-mode user interface results in highly interactive systems where users can easily switch back and forth between paper and digital information. The definition of an abstract input device interface further provides flexibility for supporting emerging technologies for paper link definition in addition to the hardware solutions for paper link definition and activation that were developed within the Paper++ project.

Claus Atzenbeck, WildDocs - Investigating Construction of Metaphors in Office Work, Aalborg University, Jul 2006, PhD.
Abstract | Full document

Knowledge is an important resource in an information society. People use it to develop new products, find new medical treatment to fight diseases, adapt or change international relationships, or gain new knowledge. The amount of knowledge constantly grows and therefore produces the need for tools that help people to structure, store, or retrieve information. One example of a tool is paper, an old medium that still is used to store, exchange, or retrieve information. People have been refining this carrier to support knowledge worker in managing their work load. Paper needs to be organized. Therefore, libraries were invented even in ancient times that allow, for example, civil servants to reach the information they need efficiently. This is still true today. In the late 1970s, another tool that aims to solve the problems of structuring, storing, and retrieving of information was introduced: personal computers. They have become normal devices in today's offices, used by millions of office workers. Computers can run databases that store large amounts of data or offer information retrieval systems that support the user in finding information. Recently, semantic technologies became popular in computer science. They allow applications (so-called "agents") to use attached semantics for improving retrieval related functionality. Another branch of research and applications focus on structure domains. There is a variety of structures, each built with well-defined tasks in mind. For example, taxonomic structures are appropriate for classification, such as those used in biology. The classification must exist before biologists can start classifying plants or animals. These techniques may not be appropriate when the final structure is not known. For example, associations come up during brainstorming sessions that do not follow a predefined structure. One way to represent the associations is to use small pieces of paper, write or draw the associated term or a picture on it, and place it on a workspace. During the session, participants can move these nodes around to express relationships among the information snippets. The structure changes constantly over time. Most of the structure is implicit, such as spatial arrangement based on completely freely movable snippets. Brainstorming sessions usually take only a short period of time. Other paper-based structures, for example, structures of printed articles or books on a desk, evolve over weeks or months. Devices, such as binders, folders, or shelves, help to put them in place. In many cases, offices have to be restructured due to lack of space or growing pieces of information. That leads to emerging spatial arrangements that exist beside predefined ones (e.g., computer science books are located at the top part of the shelve). Also here, spatial structures that were created or modified over time carry implicit metainformation to a large extent. This metainformation is best interpreted by the person who created it.

David Parry, Fuzzy Ontology and Intelligent Systems for Discovery of Useful Medical Information, AUT, Mar 2006, PhD.
Abstract

Currently, reliable and appropriate medical information is difficult to find on the Internet. The potential for improvement in human health by the use of internet-based sources of information is potentially huge, as knowledge becomes more widely available, at much lower cost. Medical information has traditionally formed a large part of academic publishing. However, the increasing volume of available information, along with the demand for evidence based medicine makes Internet sources of information appear to be the only practical source of comprehensive and up-to-date information. The aim of this work is to develop a system allowing groups of users to identify information that they find useful, and using those particular sources as examples develop an intelligent system that can classify new information sources in terms of their likely usefulness to such groups. Medical information sources are particularly interesting because they cover a very wide range of specialties, they require very strict quality control, and the consequence of error may be extremely serious, in addition, medical information sources are of increasing interest to the general public. This work covers the design, construction and testing of such a system and introduces two new concepts – document structure identification via information entropy and fuzzy ontology for knowledge representation. A mapping between query terms and members of an ontology is usually a key part of any ontology enhanced searching tool. However many terms used in queries may be overloaded in terms of the ontology, which limits the potential use of automatic query expansion and refinement. In particular this problem affects information systems where different users are likely to expect different meanings for the same term. This thesis describes the derivation and use of a "Fuzzy Ontology" which uses fuzzy relations between components of the ontology in order to preserve a common structure. The concept is presented in the medical domain. Kolmogorov distance calculations are used to identify similarity between documents in terms of authorship, origin and topic. In addition structural measures such as paragraph tags were examined but found not to be effective in clustering documents. The thesis describes some theoretical and practical evaluation of these approaches in the context of a medical information retrieval system, designed to support ontology-based search refinement, relevance feedback and preference sharing between professional groups.

Frank Allan Hansen, Context-aware Mobile Hypermedia: Concepts, Framework, and Applications, University of Aarhus, Dec 2006, PhD.
Abstract | Full document

Context-aware computing focuses on developing systems which through different kinds of sensory systems are able to collect information about the physical environment in order to, more or less automatically, adapt data and behavior to the userÂ’s current context. The aim is to create systems that, rather than solely depend on explicit user input, can react to changes in the use-context and thereby be easier to use. Contextualization of information is also inherent to hypermedia. Through links and other structuring mechanisms hypermedia has striven to support users in accessing and understanding information in different contexts. Unfortunately, hypermedia has traditionally been confined to structuring digital material, and the associations which we as humans are able to maintain between (digital) information and physical entities, have therefore largely been unsupported. But as computational power moves beyond the desktop and into the physical environments we live and work in, it may be worthwhile to support these digital-physical relationships. The thesis addresses the definition, design and requirements for context-aware mobile hypermedia systems. Context-aware hypermedia applies ubiquitous computing techniques to the classical concepts of open hypermedia in order to support linking of both digital and physical objects and to provide information adapted to the use-context. The results presented here originate from work performed within two research projects, ContextIT and the iSchool projects, and focus on support for fieldwork and project based education in Danish elementary schools and in particular, the process of collecting, producing and presenting information both in and out of the classroom. The main contribution of this work is the provision of a better understanding of how to meet the requirements for context aware mobile hypermedia, both theoretically and practically, and is realized by: A conceptual model for context-aware hypermedia, HyCon, a framework for context-aware mobile hypermedia, A range of hypermedia applications utilizing context-awareness to support mobile fieldwork.

Nor Aniza Abdullah, An Architecture for Augmenting the SCORM Run-Time Environment with a Personalised Link Service, University of Southampton, Jan 2006, PhD.
Abstract | Full document

As the result of recent advances in the business of e-learning there has been a growing interest in e-learning standards, particularly SCORM (Sharable Content Object Reference Model). SCORM is a reference model that integrates a collection of e-learning resource standards and specifications. In a SCORM compliant courseware, content and the pedagogic approach to be taken are predefined by the course author. As a consequence, users are unable to learn according to their preferences, and implicitly they will all encounter the same learning experience. Recent attempts to personalise learning in SCORM often resulted in either modifying SCORM or substituting its course sequencing mechanism with Adaptive Hypermedia techniques. Nonetheless, SCORM is a widely-used solution to interoperability problems with e-learning resources and can realise not only content sharing and reusability but also a consistent sequencing of course content across different systems and tools. For these reasons, this thesis focuses on supplementing SCORM sequencing rather than redefining it. The novelty of this work is that it builds an adaptive environment around the existing SCORM, without extending or modifying SCORM itself. The work integrates Adaptive Hypermedia principles into the SCORM Run-Time Environment as an independent service to support the environment with learning materials that are preferable to both teachers (primary materials of their choice) and students (supplementary materials that are pre-selected to suit some aspects of their user models to assist their understanding of the primary materials). This SCORM-complementing approach can also enable the SCORM Run-Time Environment to include on-demand external resources into the environment in order to address SCORM’s limitation of static pool of learning resources. The novel contribution of this work is the design of an authoring architecture which enables the automatic generation of a concept map and consequently links to alternative learning resources, and a run-time service oriented architecture which delivers these alternative resources, alongside the SCORM defined resources, according to a dynamic user model.

Renato de Freitas Bulcão Neto, Um processo de software e um modelo ontológico para apoio ao desenvolvimento de aplicações sensíveis a contexto, Universidade de São Paulo, Dec 2006, PhD.
Abstract

In order to provide adaptive services according to users’ tasks, context-aware applications exploit context information, which is any information used to characterize entities of a user-computer interaction such as user identity or user location. This thesis deals with the lack of a software process-based approach to supporting the inherent complexity of developing context-aware systems. The work reported in this thesis followed three main lines of investigation: context information modeling, services for processing context information, and a software process for context-aware computing. The contributions of this thesis include: (i) the Process for Ontological Context-aware Applications (POCAp) to support the development of context-aware applications based on ontologies; (ii) the Semantic Context Model (SeCoM) based on Semantic Web standards and ontologies; (iii) the Semantic Context Kernel (SCK) services infrastructure for interpreting ontological context information models such as the SeCoM model; (iv) an implementation of the POCAp process for the extension of an application with context information based on the SeCoM model, and its integration with services of the SCK infrastructure; and (v) the identification of design issues related to the inference over ontology-based context information.

Simon Grange, A Virtual University Infrastructure For Orthopaedic Surgical Training With Integrated Simulation, University of Southampton, Jan 2006, PhD.
Abstract | Full document

This thesis pivots around the fulcrum of surgical, educational and technological factors. Whilst there is no single conclusion drawn, it is a multidisciplinary thesis exploring the juxtaposition of different academic domains that have a significant influence upon each other. The relationship centres on the engineering and computer science factors in learning technologies for surgery. Following a brief introduction to previous efforts developing surgical simulation, this thesis considers education and learning in orthopaedics, the design and building of a simulator for shoulder surgery. The thesis considers the assessment of such tools and embedding into a virtual learning environment. It explains how the performed experiments clarified issues and their actual significance. This leads to discussion of the work and conclusions are drawn regarding the progress of integration of distributed simulation within the healthcare environment, suggesting how future work can proceed.

Stefano Bocconi, Vox Populi: Generating Video Documentaries From Semantically Annotated Media Repositories, CWI Amsterdam, Nov 2006, PhD.
Abstract | Full document

The scenario in which the present research takes place is that of one or more online video repositories containing several hours of documentary footage and users possibly interested only in particular topics of that material. In such a setting it is not possible to craft a single version containing all possible topics the user might like to see, unless including all the material, which is clearly not feasible. The main motivation for this research is, therefore, to enable an alternative authoring process for film makers to make all their material dynamically available to users, without having to edit a static final cut that would select out possible informative footage. We developed a methodology to automatically organize video material in an edited video sequence with a rhetorical structure. This was enabled by defining an annotation schema for the material and a generation process with the following two requirements: - the data repository used by the generation process could be extended by simply adding annotated material to it - the final resulting structure of the video generation process would seem familiar to a video literate user The first requirement was satisfied by developing an annotation schema that explicitly identifies rhetorical elements in the video material, and a generation process that assembles longer sequences of video by manipulating the annotations in a bottom-up fashion. The second requirement was satisfied by modelling the generation process according to documentary making and general film theory techniques, in particular making the role of rhetoric in video documentaries explicit. A specific case study was carried out using video material for video documentaries. These used an interview structure, where people are asked to make statements about subjective matters. This category is characterized by rich information encoded in the audio track and by the controversiality of the different opinions expressed in the interviews. The approach was tested by implementing a system called Vox Populi that realizes a user-driven generation of rhetoric-based video sequences. Using the annotation schema, Vox Populi can be used to generate the story space and to allow the user to select and browse such a space. The user can specify the topic but also the characters of the rhetorical dialogue and the rhetoric form of the presentation. Presenting controversial topics can introduce some bias: Vox Populi tries to control that by modelling some rhetoric and film theory editing techniques that influence the bias and by allowing the user to select the point of view she wants the generated sequence to have.

Suely de Brito Clemente Soares, CiberEduc: Building and Developing a Virtual Collaborative Learning Community of ICT Applied to the Daily Work of Brazilian Universities Reference Librarians, UNICAMP Faculty of Education, Feb 2006, PhD.
Abstract | Full document

It´s already unusual in Brazil the exploration of cyberspace for educational purposes, specially regarding to ability professionals of information through the remote long living education. After that presupposition this research was developed through qualitative approach. It can be classified at explicative level and labeled as almost experimental methodology. The purpose of this study was to build to develop and to observe the interactions at cyberspace a collaborative learning community with brazilian universities reference librarians self-interested in the ICT applied to their daily work, wich were invited to participate as subjects of this research by posted messages in their discussion lists. The proposal offered to them was spontaneous and free participation during the second semester of 2003 at a virtual community called CyberEduc. It was built at UNICAMP´s TelEduc virtual environment. A CyberEduc´s presentation video was available UNICAMP´s cameraweb to be watched via internet. The 118 active subscribers with the researcher and her adviser shaped a community of 120 cyberstudents who became from Brasilia, DF and 12 brazilian´s states. The CyberEduc was structured around an agenda that pointed out to 14 different activities. An initial and another final interview was the first as well the last activity. The analyzed and caught data offered a vision over the subject´s interactions with the environment and between themselves also the contributions gave to the CyberEduc like participating with their personal files, doubts, critics, suggestions and matters. The log number to the CyberEduc was 2081. The sum of all environment links access was 13140. It was described all discussed ICT. The chapters describe the environment tools configuration, the CyberEduc community developing and some additional contributions to the scientific knowledge about collaborative learning virtual communities. Because of quantity either quality of interactions happened, the final considerations witness that was developed, in fact, an online ICT´s collaborative learning community.

Flavius Frasincar, A Model-driven Approach for Building Distributed Ontology-based Web Applications, Eindhoven University of Technology, Jun 2005, PhD.
Abstract | Full document

Due to Web popularity many information systems have been made available through the Web, resulting in so-called Web Information Systems (WIS). Due to the complex requirements that WIS need to fulfill, the design of these systems is not a trivial task. Design methodologies provide guidelines for the construction of WIS such that the complexity of this process becomes manageable. Based on the separation-of-concerns principle some of these methodologies propose models to specify different aspects of WIS design. Model-driven WIS design methodologies have been recently influenced by emerging technologies like the ones provided by the Semantic Web. We call WIS that use Semantic Web technologies Semantic Web Information Systems (SWIS). Hera is a SWIS design methodology that employs RDF, the foundation language of the Semantic Web, for its model representation. Using a standardized language to represent models fosters application interoperability. There are two main phases in Hera: the data collection phase, which makes available data coming from different possibly heterogeneous sources, and the presentation generation phase, which builds Web hypermedia presentations based on the previously integrated data. This dissertation concentrates on the presentation generation phase of the Hera methodology. The presentation generation phase of the Hera methodology identifies the following design steps: conceptual design: constructs the conceptual model (CM), a uniform representation of the application's data. It defines the concepts and the concept relationships that are specific to the application's domain. application design: constructs the application model (AM), the navigational structure over the application's data. It defines slices and slice relationships. A slice is a meaningful presentation unit of some data. There are two types of slice relationships: navigation relationships, used for links between slices, and aggregation relationships, used for embedding a slice into another slice. presentation design: constructs the presentation model (PM), the look-and-feel specifications of the presentation. It defines regions and region relationships. A region is an abstraction for a rectangular area on the user display where the contents of a slice are presented. As for slices, there are two types of region relationships: navigation relationships, used for links between regions, and aggregation relationships, used for embedding a region into another region. Regions have associated layout (positioning of inner regions inside a region) and style (fonts, colors, etc.) information.

Jessica Rubart, THINK SHARED – Modellbasierte und komponentenorientierte Unterstuetzung wissensintensiver Kooperationen, Fern Universitaet Hagen, Feb 2005, PhD.
Abstract | Full document

Groupware aims at improving the cooperation between humans in terms of efficiency and effectiveness. In the course of knowledge intensive cooperation the behavior of the participants is significantly influenced by knowledge. It is, thus, extremely important to support shared understanding about the cooperation context. Such a context cannot be prescribed in detail; instead, it evolves over time. Cooperative visual modeling is an important means for supporting shared understanding. Compared to existing groupware approaches this thesis presents a holistic and infrastructural approach aiming at model-based support for knowledge intensive cooperation. In addition, it shows a technical implementation. Holistic means that this approach integrates modeling, using, and post processing relevant activity knowledge. Infrastructural means that this approach can be configured and extended (where appropriate) for a specific usage scenario. Such a holistic, infrastructural approach improves efficiency and effectiveness of knowledge intensive cooperation. In this thesis, different knowledge intensive cooperation scenarios are analyzed and requirements for a supporting infrastructure are derived. In particular, cooperative meta-modeling and flexibility of models are important. In order to fulfill the identified requirements this thesis presents the approach THINK SHARED. It is characterized by a cooperative knowledge management cycle, which flexibly integrates cooperative meta-modeling, modeling, and using knowledge, which represents the cooperation context. THINK SHARED presents shared hypermedia workspaces that integrate cooperative modeling of the cooperation context and usage of shared knowledge models. Furthermore, solutions for cooperative configuration and extension of such shared hypermedia workspaces are developed addressing cooperative meta-modeling and extensibility. In particular, component technology and component-oriented development are important. The latter includes the development for reuse and with reuse and constitutes another methodological element of the approach. The system XCHIPS4KM is a concrete implementation of the approach and described in the thesis. This prototype and previous versions have been used for supporting different knowledge intensive cooperation scenarios. The presented implementation, usage experiences, and comparison with related work show that THINK SHARED is practical, useful, and novel. Moreover, the approach can be applied to many different research and application domains.

Pablo Cesar, A Graphics Software Architecture for High-End Interactive TV Terminals, Finland University of Technology, Dec 2005, PhD.
Abstract | Full document

This thesis proposes a graphics architecture for next-generation digital television receivers. The starting assumption is that in the future, a number of multimedia terminals will have access through a number of networks to a variety of content and services. One example of such a device is a media station capable of integrating different kinds of multimedia objects such as 2D/3D graphics and video, reacting to user interaction, and supporting the temporal dimension of applications. Some of the services intended for these devices include, for example, games and enhanced information over broadcasted video. First, this thesis provides an overview of the digital television environment, focusing on the limitations of current receivers and hints at future directions. In addition, this thesis compares different solutions from regional standardisation bodies such as DVB, CableLabs, and ARIB. It proposes the adoption of the most relevant initiative, GEM by DVB. Unfortunately, GEM software middleware only considers Java language as an authoring format, meaning that the declarative environment and advanced functionalities (e.g., 3D graphics support) remain to be standardised. Because in the future different user groups will have different demands with regard to television, this thesis identifies two major extensions to the GEM standard. First, it proposes a declarative environment for GEM that takes into account W3C standardisation efforts. This environment is divided into two configurations: one capable of rendering limited interactive applications such as information services, and another intended for more demanding applications, for example a distance learning portal that synchronises videos of lecturers and slides. Second, this thesis proposes to extend the procedural environment of GEM with 3D graphics support. The potential services of this new profile, High-End Interactive, include games and commercials. Then, based on the requirements the proposed profiles should meet, this thesis defines a graphics architecture model composed of five layers. The hardware abstraction layer is in charge of rendering the final graphics output. The graphical context is a cross-platform abstraction of the rendering region and provides graphics primitives (e.g., rectangles and images). The graphical environment provides the means to control different graphical contexts. The GUI toolkit is a set of ready-made user interface widgets and layout schemes. Finally, high-level languages are easy-to-use tools for developing simple services. The thesis concludes with a report of my experience implementing a digital television receiver based on the proposals described. In addition to testing the application of the proposed graphics architecture to the design and implementation of a next-generation digital television receiver, the implementation permits the analysis of the requirements of such receivers and of the services they can provide.

Richard Vdovjak, A Model-driven Approach for Building Distributed Ontology-based Web Applications, Technical University Eindhoven, Jun 2005, PhD.
Abstract | Full document

The overwhelming success of the World Wide Web has caused a change in the notion of Information Systems by adopting the Web paradigm both as the delivery platform and also as the source of their data. The Semantic Web initiative opens new possibilities which the "old" World Wide Web could not deliver. It also brings, however, new set of requirements for information system design. Designing an information system for the Semantic Web requires the use of a thorough design methodology. We propose Hera, a model-driven methodology supporting WIS design, focusing on the processes of integration, data retrieval, and presentation generation. All Hera models are based on RDF(S), the pivot language of the Semantic Web, making the methodology a suitable candidate for designing true Semantic Web Information Systems. When the content of such a system is gathered from different information sources, the specification of how data is to be retrieved requires an appropriate specification framework and a suite of tools which are able to process the designed specifications and retrieve the data as a respons to the user query. The contribution of this thesis lies in designing the general Hera framework and in particular, providing a solution to the problem of how to specify the outline of integration of heterogeneous information sources on the Semantic Web in order to facilitate a uniform access to their distributed data. This carefully designed view of selected pieces of data coming from different sources constitutes the semantic layer of the Hera suite, which is subsequently used by other modules to deliver a tailored hypermedia presentation to the enduser. To be able to to specify the semantic layer we propose an Integration Model formalism which is able to deal with many semantic heterogeneities that frequently occur among sources on the Semantic Web. We designed and implemented a prototype of the integration engine which serves as a backend of the Hera suite providing the semantic layer for the rest of the framework. Due to the fact that the proposed Hera architecture is modular with clearly defined interfaces built on top of the RDF foundation, the use of the integration engine is not limited solely for the Hera suite. It can also serve as a stand-alone general purpose distributed RDF query engine allowing other parties to query the semantic layer of the designed application, dus potentially extending the content available on the Semantic Web. As the integration engine is often faced to process large amounts of RDF data, the performance becomes an important issue. Several optimization techniques were adopted to improve the speed of query processing; these include specially tailored index structures for intelligent query routing, join ordering techniques for efficient result assembly, and algebraic optimizations for improving query plans; for this a special purpose RDF algebra (RAL)was proposed.

Yeliz Yesilada, Annotation and Transformation of Web Pages to Improve Mobility for Visually Impaired Users, University of Manchester, Aug 2005, PhD.
Abstract | Full document

In the last decade, the growth and the popularity of the World Wide Web (Web) have been phenomenal. Originally, it was a purely text-based system that allowed assistive technologies to be designed to transform pages into alternative forms (e.g., audio) for disabled people. This meant that for the first time, a vast amount of information was available and easily accessible to disabled people. However, advances in technologies and changes in the main authoring language, transformed the Web into a true visualcommunication medium. These changes eventually made the Web inaccessible to visually impaired users. In particular, travelling around the Web became a complicated task, since the richness of visual navigational objects presented to their sighted counterparts are neither appropriate nor accessible to visually impaired users. This thesis investigates principles and derived techniques to enhance the travel experience for visually impaired Web users. The hypothesis is that travel support for visually impaired users can be improved if Web pages are analysed to identify the objects that support travel and are then transformed in such a way that they can then fulfill their intended or implicit roles. This hypothesis is supported by the identification of structural and navigational properties of Web pages which have been encapsulated into an ontology (WAfA) to support machine processing; and the design of a flexible pipeline approach to annotate and transform Web pages by using this ontology. An experimental test-bed, Dante, has also been created based on this pipeline approach, that encodes these principles and techniques to transform Web pages. Finally, a user evaluation method is devised and applied to demonstrate that the travel experience of visually disabled users can be improved through the application of these techniques. This research demonstrates that by providing machine processable data, regarding the structural and navigational properties of Web pages, applications can be created to presentWeb pages in alternative forms and so enhance the travel experience of visually impaired users. The work presented in this thesis is of practical value to the Web accessibility community and is an important case study for demonstrating Semantic Web technologies. By moving away from thinking that simple translation of text to audio is enough to provide access to Web pages, this thesis is part of the coming of age of assistive technology and is a significant contribution to redressing the inequality arising from visual dominance in the Web.

Susana Montero, Integration of Patterns Into Design Process of Hypermedia Systems by Means of Ontologies, Carlos III University, Jul 2005, PhD.
Abstract

The development of large-scale hypermedia and web systems must be carried out following a well defined systematic process, that is, applying hypermedia and web engineering methods and techniques. In addition, considering that the development process of web systems is characterized by obtaining an operative product in a short period of time, designers can alleviate decision making resorting to design patterns. A design pattern records the knowledge and the experience of domain experts on how to make a software system more reusable and flexible as a result of many efforts on the design and codification of systems. Currently, if a designer or a developer of hypermedia applications wants to solve or to communicate a specific design problem by means of hypermedia design patterns (hdp), as starting point, she would have to search across multiple existing publications or some of web repositories available in order to identify the pattern or patterns that best fit her needs. To date, the necessary mechanisms to help the designer to find the appropriate pattern or patterns and to integrate them into hypermedia design methods do not exist. Therefore, the success of this activity will depend only on the designer knowledge about the existing resources and her experience in the use of patterns. In order to solve these lacks, the first aim of this thesis has been to organize the knowledge of hdp, providing a set of classification criteria which takes into account as much hypermedia application modeling needs as the different levels of abstraction used in the pdh literature. From these criteria, a pattern catalogue has been elaborated to facilitate the search of patterns during the resolution of particular design problems as well as a pattern language as a step-by-step guide that advises designers how to accomplish the design process from different levels of abstraction, whereas it proposes alternative problems to resolve. Finally, from the pattern language, an integration framework with design methods has been specified, in which a set of steps guides a pattern based approach to transform requirements specifications to design entities into the conceptual modeling, thanks to the solutions captured by patterns. This process has been evaluated in an empirical way to verify its validity and utility.

Jing Zhoun, DDLS: Extending Open Hypermedia Systems into Peer-to-Peer Environments, University of Southampton, Jan 2004, PhD.
Abstract | Full document

Peer-to-peer (P2P) computing is primarily characterised by decentralisation, scalability, anonymity, self-organisation and ad hoc connectivity. It attracted considerable attention in open hypermedia research due to its potential for supporting collaboration among a community of people sharing similar knowledge background. The aim of this research is to investigate the feasibility and potential benefits of corporating the P2P paradigm in open hypermedia systems to support resource sharing-based collaboration. This is accomplished by utilising a distributed dynamic link service (DDLS) as a testbed, addressing issues that arise from implementing the paradigm, and demonstrating the efficiency of proposed techniques through simulation. This research begins with the development of a prototype DDLS using the open hypermedia paradigm for storing and presenting resources and a centralised P2P model which adopts a central service directory for publishing and discovering resources in a well-arranged environment. This is enhanced by an operational analysis and feature comparison between prototypes based on the traditional client-server and the centralised P2P models. Various P2P models are analysed to identify the key characteristics of and requirements for the DDLS using an unstructured P2P model which empowers collaboration in an ad hoc environment. The second phase of this research concentrates on overcoming the challenges of resource description, publishing and discovery posed by the unstructured P2P DDLS: using RDF to encode information about resources, developing a clustering technique to group resources and form the information space; and creating a semantic search mechanism to discover resources; respectively. Finally, this research proposes re-organisation techniques based on the exponential decay function and the naive estimator to enhance the performance of resource discovery in resource sharing-based collaboration.

Owen Conlan, The Multi-Model, Metadata Driven Approach to Personalised eLearning Services, Trinity College, Dublin, Nov 2004, PhD.
Abstract | Full document

One of the major obstacles in developing quality eLearning content is the substantial development costs involved and development time required [Marchionini, 95]. Educational providers, such as those in the university sector and corporate learning, are under increasing pressure to enhance the pedagogical quality and technical richness of their course offerings while at the same time achieving improved return on investment. One means of enhancing the educational impact of eLearning courses, while still optimising the return on investment, is to facilitate the personalisation and repurposing of learning objects across multiple related courses. However, eLearning courses typically differ strongly in ethos, learning goals and pedagogical approach whilst learners, even within the same course, may have different personal learning goals, motivations, prior knowledge and learning style preferences. This thesis proposes an innovative multi-model approach to the dynamic composition and delivery of personalised learning utilising reusable learning objects. The thesis describes a generic and extensible adaptive metadata driven engine that composes, at runtime, tailored educational experiences across a single educational content base. This thesis presents the theoretical models, design and implementation of an adaptive hypermedia educational service. It also describes how this multi-model, metadata driven approach, and the adaptive engine built in accordance with this approach, was trialled and evaluated from pedagogical, reusability and technical perspectives.

Patrick Sinclair, Integrating Hypermedia Techniques with Augmented Reality Environments, University of Southampton, Jan 2004, PhD.
Abstract | Full document

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.

Timothy Miles-Board, Everything Integrated: A Framework for Associative Writing in the Web, University of Southampton, Jan 2004, PhD.
Abstract | Full document

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.

Anders Fagerjord, Rhetorical Convergence: Earlier Media Influence on Web Media Form, University of Oslo, Nov 2003, PhD.
Abstract | Full document

This thesis takes as its starting point the widespread conception of media convergence. If all media are converging, as many believe, what do the texts look (or sound) like in the new convergent media? Using feature journalism on the Web as the object of analysis, Fagerjord asks whether the means of expression of earlier media are converging on the Web, and if so, how. Fagerjord argues that many Web genres can be viewed as cases of rhetorical convergence, following various forms of technological convergence of digital media. Using semiotic and rhetoric theory, a large number of Web sites and genres are analysed. The thesis consists of nine essays, each addressing a specific problem related to the multimediated composition of Web sites. More theoretical problems include questions of linearity and nonlinearity in Web sites, and the use of semiotic theory on computer texts. Specific examples of multimedia genres are analysed in detail, such as interactive graphics, photo collections on the Web, Web sites combining video with written material, and "edu-tainment" texts resembling computer games. In the introductory essay, the insights from the following contributions are brought together to a higher and more general level of analysis, as elements in a four-dimensional theory of relations between Web genres and earlier media. Fagerjord argues that any text may be described according to the four axes Mode of Distribution (the balance of amount of material and time between authoring and reading); Mode of Restrictions (range and detail in space and time); Mode of Acquisition (the reading process required of the reader); and Mode of Signification (the particular combination of sign systems). Rhetorical convergence is when a text is similar to one genre on one axis and another genre on another axis. However, the model implies that rhetorical divergence may be better than convergence in describing the changes.

Jianhan Zhu, Mining Web Site Link Structures for Adaptive Web Site, University of Ulster, Oct 2003, PhD.
Abstract

This thesis is concerned with mining the log file of a Web site for knowledge about the Web site and its users, and using the knowledge to assist users to navigate and search the Web site effectively and efficiently. First, we investigate approaches to adapting the organization and presentation of a Web site by learning from the Web site link structure and user behavior of the Web site. Approaches are developed for presenting a Web site using a link hierarchy and a conceptual link hierarchy respectively based on how users have used the Web site link structure. Link hierarchies and conceptual link hierarchies can be used to help users navigate the Web site. Second, we develop approaches for building a first-order Markov chain model of user navigation on the Web site link structure, link hierarchy, and conceptual link hierarchy respectively. Under a collaborative assumption, the model can be used for link prediction that assists users to navigate the Web site. Third, approache's are developed for ranking Web pages based on how users have used the Web site link structure. The page rankings can be used to help users search the Web site. The approaches developed in the thesis have been implemented in a prototype called Online Navigation Explorer (ONE). First, link hierarchies and conceptual link hierarchies are visualized in ONE. Second, link prediction using Markov chain models is integrated with link hierarchies and conceptual link hierarchies in ONE. Third, search results are visualized in ONE. Experimental results show that ONE can help users navigate a Web site and search for their desired information on the Web site effectively and efficiently. The work presented in the thesis is a step towards the development of an adaptive Web site, which can assist users to navigate the Web site and search for their desired information on the Web site.

Christopher P. Bailey, An Agent-Based Framework to Support Adaptive Hypermedia, University of Southampton, Jan 2002, PhD.
Abstract | Full document

The field of adaptive hypermedia is a little over a decade old. It has a rich history in a range of fields such as artificial intelligence, user modelling, intelligent tutoring systems and hypertext. Early adaptive hypermedia work concentrated on application-led research; developing a range of systems for specific purposes. In 1996, Peter Brusilovsky reviewed the state-of-the-art and proposed a taxonomy of adaptive hypermedia techniques, thereby providing the means to categorise adaptive hypermedia systems. Since then, several practical frameworks for adaptive hypermedia applications have been produced, in addition to formal models for formalising adaptive hypermedia applications. This thesis presents a new framework for adaptive hypermedia systems based on agent technology, a field of research largely ignored within the adaptive community. Conceptually, this framework occupies a middle ground between the formal reference models for adaptive hypermedia and application-specific frameworks. This framework provides the means to implement formal models using variety of architectural approaches. Three novel adaptive hypermedia applications have been developed around this agent-based framework. Each system employs different architectural structures, they model the user with a variety of profiling techniques, and each provides a different set of adaptive features. The diversity of these three systems emphasises the flexibility and functionality of this proposed agent-based framework.

Simon Kampa, Who Are The Experts? e-Scholars in The Semantic Web, University of Southampton, Jan 2002, PhD.
Abstract | Full document

Scholarly research is the sum of diverse activities and leads to the dissemination of a large amount of material. Traditional approaches to exploring and becoming proficient within an esoteric research field rely on slow and sometimes ineffective discourse, and depend more on a scholar's detective skill, effort, and perseverance. However, the Web has introduced the potential for improved accessibility, interconnectivity, and more efficient and effective communication; we are becoming e-Scholars. Current efforts on the Web have focussed mainly on improving the accessibility of on-line research material and as a result, researchers have yet to realise the full implications of the new medium. Consequently, the emphasis must shift to improving and enhancing access to scholarly material; this research proposes a novel approach by presenting researchers with the facility to comprehensively, extensively, and rationally explore their research field and ask intricate questions about it and the individual facts and issues raised by it. This is realised through the integration of principles from the hypertext, Semantic Web, and digital library fields to interconnect and analyse all scholarly material in the academic domain. The e-Scholar Knowledge Inference Model (ESKIMO) demonstrates the approach and provides a platform for evaluation and further research.

Anja Rau, What You Click is What you Get? - Die Stellung von Autoren und Lesern in Interaktiver Digitaler Literatur, Johannes Gutenberg-Universität Mainz, Jan 2000, PhD.
Abstract

In the 1990s, literary hypertext was lauded for its postmodern ability to transform the reader into an active and creative co-author. George Landow's "almost embarassingly literal embodiment" hypothesis seemed to promise a revolutionary paradigm shift in the way we create and receive literature. At the same time, scholars and critics began to call for "real criticism of real hypertext". This doctoral thesis offers in-depth analyses of ten hyperfictions and 18 adventure games and focuses on the position of authors and readers in interactive digital fiction. It arrives at the conclusion that the physical interaction the computer offers to readers does not really activate the reader into an author-like position of creativity. The mental interaction of the reader with the digital text, on the other hand, is indeed of a different quality than the interaction with a printed text, though not necessarily better, or more empowering. In fact, the digital environment serves to increase the authority of the author.

David E. Millard, Hypermedia Interoperability: Navigating the Information Continuum, University of Southampton, Jan 2000, PhD.
Abstract | Full document

Open Hypermedia Systems are designed to allow links to be authored and followed on top of any media format. The link structures are held separately from the documents in a software component called a Link Server. As hypermedia has matured as a research topic attention has turned to standardising the way in which components talk to Link Servers in order to provide interoperability. The Open Hypermedia Systems Working Group took up this challenge and proposed an Open Hypermedia Protocol (OHP). However, the scope of this proposal proved to be too large and the protocol was divided into domain specific parts (Navigational, Spatial and Taxonomic Hypermedia), tackling each domain differently, but consistently. It is questionable whether this step was the correct one, as the domains share many similar features. In this thesis I present a detailed examination of the information spaces that the OHP was attempting to model (from all these considered hypertext domains), which incorporates notions of both behaviour and context. This examination looks at what it means to navigate around the many dimensions of information, across these domains, and reveals a cohesive and continuous structure that I call the Information Continuum. The Fundamental Open Hypermedia Model (FOHM) is presented, which is capable of representing the structures of this continuum in a consistent and meaningful way. FOHM is coupled with an agent infrastructure to produce an implementation that demonstrates the model being used for cross-domain interoperability.

Gary Wills, The Design and Evaluation of Industrial Hypermedia, University of Southampton, Jan 2000, PhD.
Abstract | Full document

This thesis proposes design, authoring and evaluation methodologies to address the issues faced by an enterprise when developing a large-scale industrial strength hypermedia application, for use in a manufacturing environment. The design methodology satisfies the requirements found in the industrial environment in a holistic approach. That is, the methodology takes into account the information structures, the constraints that exist within the industrial environment, and the background of the people who will have to implement the hypermedia application. The size and complexity of an industrial environment aggravates the problems found in general hypermedia authoring, especially in the areas of scalability, re-use and cognitive overload to the author. Hence, an authoring methodology is present, which addresses these issues. The design and authoring methodologies are evaluated against an independent set of criteria. The results show, even after sensitivity analysis, that when compared against other methodologies, the methodologies presented in this thesis are the ones better suited to meet the requirements of an industrial application. The design and authoring methodologies have been used to produce an industrial hypermedia application. The application communicated with existing information systems, factory floor equipment and the management network. The application has been used to evaluate the usability of the system within the factory environment. A cost effective user evaluation methodology was developed. Not only did it allow potential problems to be identified early in the project, but it also encouraged ownership of the project from the managers, maintainers and factory-floor operators. The user evaluation showed that, compared to an existing paper based system, the user preferred the hypermedia system and that they found the hypermedia application quicker and easier to use. While the majority of problems to be overcome, when introducing industrial strength hypermedia, are technical, some of the most difficult issues relate to the inertia of the management culture within the organisation. Hence, the strategic benefits and considerations when introducing hypermedia application into an organisation are discussed. In addition a model for estimating the effort to produce an industrial hypermedia application, has been developed

Jacco van Ossenbruggen, Processing Structured Hypermedia: A Matter of Style, Vrije Universiteit Amsterdam, Jan 2000, PhD.
Abstract | Full document

With the introduction of the World Wide Web in the early nineties, hypermedia has become the uniform interface to the wide variety of information sources available over the Internet. The full potential of the Web, however, can only be realized by building on the strengths of its underlying research fields. This book describes the areas of hypertext, multimedia, electronic publishing and the World Wide Web and points out fundamental similarities and differences in approaches towards the processing of information. It gives an overview of the dominant models and tools developed in these fields and describes the key interrelationships and mutual incompatibilities. In addition to a formal specification of a selection of these models, the book discusses the impact of the models described on the software architectures that have been developed for processing hypermedia documents. Two example hypermedia architectures are described in more detail: the DejaVu object-oriented hypermedia framework, developed at the VU, and CWI\'s Berlage environment for time-based hypermedia document transformations.

Jim Whitehead, An Analysis of the Hypertext Versioning Domain, University of California, Irvine, Jan 2000, PhD.
Abstract | Full document

Hypertext captures the implicit and explicit relationships between intellectual works, storing them as data items within the computer, thus allowing them to be navigated, analyzed, and visualized. The evolution of information artifacts such as software development projects, large document collections, and collections of laws and regulations is characterized both by change to the works and their relationships, and the desire to record this change over time. Hypertext versioning is concerned with storing, retrieving, and navigating prior states of a hypertext, and with allowing groups of collaborating authors to develop new states over time. Several systems provide hypertext versioning services; this dissertation provides a domain model of these systems, comprised of domain terminology, a taxonomy, reference requirements, a data modeling model, and design spaces associated with the requirements. This work offers several significant contributions. It provides a systematic organization of the preponderance of information concerning hypertext versioning systems, including the first taxonomy of such systems, and a comprehensive collection of their requirements. A detailed model of containment is provided; its use highlights that containment is inherent in hypertext systems, and a full understanding of hypertext versioning system data models requires an understanding of their containment relationships. The containment model allows the similarities and differences in hypertext versioning systems to be examined in a consistent manner. The design space for persistently recording revision histories employs a three-layer model that separates the abstract notion of revision history, shared by all state-based approaches, from the high-level overview of each versioning approach, which is in turn distinct from its specific concrete representation. The design space for link versioning is shown to be an application of the three-layer model for versioning works. Building on the containment model, and the design spaces for versioning works and links, the structure design space concisely describes a range of techniques for recording the history of hypertext structures. Parameters of the structure design space include the abstractions contained within the structure container, the versioning design space choice for each versioned abstraction, the containment choice for each container/containee pair, and the location of any revision selection rules.

Roy Thomas Fielding, Architectural Styles and the Design of Network-based Software Architectures, University of California, Irvine, Jan 2000, PhD.
Abstract | Full document

The World Wide Web has succeeded in large part because its software architecture has been designed to meet the needs of an Internet-scale distributed hypermedia system. The Web has been iteratively developed over the past ten years through a series of modifications to the standards that define its architecture. In order to identify those aspects of the Web that needed improvement and avoid undesirable modifications, a model for the modern Web architecture was needed to guide its design, definition, and deployment. Software architecture research investigates methods for determining how best to partition a system, how components identify and communicate with each other, how information is communicated, how elements of a system can evolve independently, and how all of the above can be described using formal and informal notations. My work is motivated by the desire to understand and evaluate the architectural design of network-based application software through principled use of architectural constraints, thereby obtaining the functional, performance, and social properties desired of an architecture. An architectural style is a named, coordinated set of architectural constraints. This dissertation defines a framework for understanding software architecture via architectural styles and demonstrates how styles can be used to guide the architectural design of network-based application software. A survey of architectural styles for network-based applications is used to classify styles according to the architectural properties they induce on an architecture for distributed hypermedia. I then introduce the Representational State Transfer (REST) architectural style and describe how REST has been used to guide the design and development of the architecture for the modern Web. REST emphasizes scalability of component interactions, generality of interfaces, independent deployment of components, and intermediary components to reduce interaction latency, enforce security, and encapsulate legacy systems. I describe the software engineering principles guiding REST and the interaction constraints chosen to retain those principles, contrasting them to the constraints of other architectural styles. Finally, I describe the lessons learned from applying REST to the design of the Hypertext Transfer Protocol and Uniform Resource Identifier standards, and from their subsequent deployment in Web client and server software.

W. J. Blustein, Hypertext Versions of Journal Articles: Computer-aided Linking and Realistic Human-based Evaluation, University of Western Ontario, Jan 1999, PhD.
Abstract

My overall objective is to develop and evaluate ways of automatically incorporating hypertext links into pre-existing scholarly journal articles. I describe a rule-based approach for making three types of links (structural, definition, and semantic). Structural links are a way of making explicit some connections between parts of the text. Definition links connect the use of a term, defined elsewhere in the document, to that definition. Links that connect parts of text that discuss similar things are semantic links. I distinguish several types of semantic links. I use two information retrieval (IR) systems (Cornell\'s SMART system and Bellcore\'s Latent Semantic Indexing) to select links based on the content of the articles. I conducted an experiment to compare the performance of the links forged using these two systems. The effectiveness of the links (and the rules used to make them) is tested by people reading the hypertext versions for information under a time constraint. A within-subjects experimental design was used. Each of the nineteen experimental participants read one version of each of three scholarly articles in a different hypertext form (one had only simple links, the others had definition links and semantic links selected using one of the IR systems). Subjects\' preferences were also measured. Although I used three survey articles from published sources for my evaluation experiment there was no difference in reader preference or performance on the basis of article. Subjects ratings of the utility of the various links shows a significant preference for structural links over semantic links. Definition links were preferred to structural links, although the result was not significant. No difference between the links created using the two IR systems was detected. However there were significant differences in the times that readers spent on documents created using the various treatments When they read in documents with only structural links readers were more likely to have read the whole article, and their satisfaction scores were inversely proportional to their comprehension score. The method of evaluating hypertext versions of journal articles for use by researchers may be applied to other hypertext versions.

Lynda Hardman, Modelling and Authoring Hypermedia Documents, University of Amsterdam, Mar 1998, PhD.
Abstract | Full document

Hypermedia presentations are documents which are not printed on paper but make use of a computer screen for their display. These include text, video, images and audio combined together as well as choice points where a reader can select other presentations to view. We use the term media items for the pieces of text, video etc. and call the choice points links. When combining media items into a presentation, temporal relations among the items specify when each item should appear on the screen and for how long. A hypermedia presentation can be generated at play-back time from an underlying document which specifies the various aspects of the presentation. To allow presentations to be played on different software systems, a model of the underlying document is needed. Another advantage of having an explicit model is that documents can be processed for other reasons, such as creating multiple versions for different end-user platforms, or changing the visual styles of the document. It is this degree of compatibility and processability which we seek to achieve with a model of hypermedia documents. Given a model for hypermedia documents, an authoring system can be created to support the creation, editing and deletion of the constituent parts of documents. This thesis first states the requirements and defines a document model for hypermedia. It then analyses the user interfaces in existing authoring systems for multimedia documents and goes on to state the requirements for a complete hypermedia authoring environment. Finally, it describes the CMIFed authoring system, implemented by members of the CWI multimedia group.

Kenneth M. Anderson, Pervasive Hypermedia, University of California, Irvine, Jun 1997, PhD.
Abstract

The heterogeneity of modern computing environments contributes to the information overload experienced by users. Relationships within and between applications, documents, and processes are often implicit and must be managed and tracked by the user. Hypermedia has been put forth as one approach to organizing these relationships, making them explicit so they can be managed. One approach to providing environment-wide hypermedia services is through the use of open hypermedia systems (OHSs). OHSs are open with respect to the set of systems and information over which hypermedia services can be provided. This research area contrasts with the original approach to hypermedia services that involved developing monolithic systems with a closed set of supported data types (e.g. HyperCard). Given the existence of OHSs, another area of research is developing integration techniques such that applications that existed before the introduction of an OHS can take advantage of the OHS’s hypermedia services. This dissertation provides contributions in both of these research fields. In particular, this work demonstrates techniques which enable OHSs to address the heterogeneity of their computing environments, to leverage the strengths of the World Wide Web (while providing the Web with improved hypermedia services), and to integrate large classes of applications at once. Heterogeneity is handled via a set of flexible abstract hypermedia concepts, application program interfaces in multiple programming languages, support for multiple operating systems, and a low entry barrier to use provided by an architecture designed to reduce the responsibilities of client applications. Integration with the Web is enabled via a scalable architecture for OHSs that is compatible with the Web’s architecture and takes advantage of the strengths of Web protocols and the familiarity of Web interaction styles. The integration of multiple applications occurs via a technique for making user-interface toolkits (and hence their constructed applications) clients of an OHS. This technique provides consistent inter- and intra-application hypermedia services. The dissertation is validated by examining the characteristics of the clients integrated with the exploratory systems developed during the course of this research. The dissertation concludes by positioning this work within the context of large-scale information environments.

Gene Golovchinsky, From Information Retrieval to Hypertext and Back Again: The Role of Interaction in the Information Exploration Interface, University of Toronto, Nov 1996, PhD.
Abstract

This work explores the design space of user interfaces for large-scale full-text database retrieval systems. Research suggests that elements of hypertext interfaces may be merged with traditional information retrieval (IR) algorithms to produce flexible hybrid interfaces for user-directed information exploration. This work examines the effectiveness of multiple-view newspaper-like interfaces, and describes a prototype that uses newspaper-style layouts to organize information retrieval results. Finally, it explores some possible visualization techniques designed to aid browsing performance. The first of two experiments in this thesis examines the effectiveness of the simultaneous display of several documents retrieved by a given query. Experimental results suggest that viewed recall increases with increasing numbers of articles displayed on the screen simultaneously. Subjects' decision-making strategies appear to be independent of user interface factors. The second experiment tests differences in behavior between query-based and link-based browsing. Differences in performance are found between groups of users employing different strategies, but not between interface conditions. These results suggest that dynamic query-mediated hypertext interfaces are viable alternatives to more explicit queries, and that subjects' intrinsic strategies have significant impact on their interaction with the system and on their performance. This work proposes an implementation of dynamic links in the WWW medium. It concludes with a discussion about the nature of hypertext interfaces and about the role of the user interface in information exploration tasks, and suggests some avenues for future research in this area.

H. Davis, Data Integrity Problems in an Open Hypermedia Link Service, University of Southampton, Jan 1995, PhD.
Abstract | Full document

A hypermedia link service is system which stores the information describing hypertext links in a database which is separate from the data content over which the links are intended to operate. One of the first open hypermedia link services was Microcosm, which takes this philosophy to the extreme, storing not only the links in a separate database, but also the information about the endpoints of the links. The most important advantage of such an organisation is that the system remains open so that hypertext functionality may be extended to third party applications. The first part of this thesis describes the background to open hypermedia link services and describes the Microcosm system, which was developed by the Multimedia Research Group at the University of Southampton. The major problem with storing all the information about links separately from the content is that such a scheme introduces many opportunities for the introduction of inconsistencies and the loss of integrity of the hypermedia data model. The second part of this thesis examines these problems, and proposes a number of solutions. It concludes that no one solution can resolve all the problems, and that in order to ensure integrity it is necessary to impose some conditions which limit the degree of openness.

L. A. Carr, Structure in Text and Hypertext, University of Southampton, Jan 1995, PhD.
Abstract | Full document

Hypertext techniques are now beginning to be used in the ways that early researchers anticipated, from personal note taking to online help for ubiquitous computing environments and universal literature resources, yet despite this, hypertext models have remained substantially unchanged. This thesis investigates the nature of text, how it may best be modelled on a computer and how the connections between related texts may be expressed in a flexible and efficient way. First we look at the development of hypertext systems and then compare that with the complex structured nature of texts themselves. LACE, a small-scale hypertext system based on structured texts is introduced. and compared with other hypertext systems. Approaches to large-scale distributed hypertexts discussed., and LACE-92, a system used to produce hypertexts from distributed information services is presented. Finally LACE-93, a new document architecture for global hypertext environments is proposed.

Weigang Wang, Semantic Net Based Hypertext for Authoring and Reuse, University of Liverpool, Jul 1995, PhD.
Abstract

The study of hypertext covers a multidisciplinary scientific and technological domain, and addresses the theories, principles, techniques and tools to structure, manage and use information in electronic format. The modularity of information units and the flexibility in organising these units make hypertext also appropriate for document reuse. A conceptual framework developed for software reuse is applied to document reuse. The framework describes reuse in terms of the processes involved. The focus of the research is hypertext methodologies and tools that support the document reuse processes, which include document creation, document management, document retrieval, and document reorganisation. Three prototype hypertext systems have been designed and implemented progressively. In each prototype, models were modified and new engineering components were added. In this way a novel system was systematically built. For the first prototype, an unstructured semantic net is exploited and an authoring tool is provided. The prototype uses a knowledge-based traversal algorithm to facilitate document reorganisation. The structural and semantic consistency of the semantic net is fully left in the users’ hands. The results indicate that the traversal method has not provided a practical solution for document reorganisation. The useful output from the traversal method depends on meaningful link and node typing. Such typing is hard to achieve with an unstructured semantic net. Then, a semi-structured semantic net was exploited and a collaborative hypertext system was built. A hierarchical view generating method was developed. Although the system suggests a small set of link types for different structures, the structural and semantic consistency of the semantic net is again left in the users’ hands. The system has proven to be useful for authoring and reusing conventional documents in a tree structure. Authors, however, were not practically able to produce hypertext that had links other than those in conventional documents in a tree structure.

MSc Theses

Hamman W. Samuel, Content Management for Online Health Advice Sharing, University of Alberta, Sep 2009, MSc.
Abstract | Full document

In the past, health information was shared by word of mouth or written down in books. With the advent of the information age, a new model has been emerging where health information consumers like patients, academicians, researchers, and doctors are turning to the Internet for health advice. With this comes questions regarding the credibility, safety, accuracy and ultimately trustworthiness of the information. These issues are particularly important because health information has the potential of affecting lives and causing harm. In this study, we highlight the distinct challenges faced by health information websites. We focus on two unique requirements: ethics and human-computer interaction. The development of a Health Content Management System (HCMS) framework that encapsulates the needs of health information websites is also discussed thoroughly. Peer-reviewed papers that contain outcomes of our research are the main components of this thesis.

Isra Al-Turaiki, An Arabic Fuzzy Expert System Shell, King Saud University, Jan 2008, MSc.
Abstract

This thesis describes the development of a fuzzy expert system shell written for Arab users. The purpose of this shell is to provide Arab users whether students or researchers with the ability to develop fuzzy expert systems using a simple Graphical User Interface. The shell is regarded as a general purpose shell.

Darren Lunn, SADIe: Structural-Semantics for Accessibility and Device Independence, University of Manchester, Sep 2005, MSc.
Abstract | Full document

Currently the World Wide Web is visual-centric with web sites often being designed only with the presentation of data in mind. A consequence of this design perspective is that information contained within the data is only accessible implicitly through the layout of the web page, rather than explicitly through the data itself. While this implicit knowledge is relatively easy to access for sighted users, it is often difficult to access for visually impaired computer users. This project describes an investigation into a way of allowing visually impaired computer users the same access to information on the World Wide Web as sighted computer users. By using ontologies to capture the semantics of the CSS Stylesheets and XHTML, the implicit information contained within a web page can be reordered and manipulated into an explicit form that better suits the needs of visually impaired users.

Frank Allan Hansen & Bent Guldbjerg Christensen, XLink-Bridging Open Hypermedia and the World Wide Web, University of Aarhus, Jan 2003, MSc.
Abstract | Full document

The World Wide Web is the single most successful hypermedia system ever developed. Based on a simple linking mechanism, the Web interconnects a massive amount of information distributed on Web servers throughout the Internet. Though very successful, the linking model has several limitations making it difficult to support knowledge work and collaboration between groups of users in a Web context. The introduction of the World Wide Web ConsortiumÂ’s XML standard and the XML linking standard, XLink, holds the promise of introducing novel approaches to the handling of and interaction with information on the Web. This thesis investigates the merits of the XLink linking language in comparison with similar advanced linking mechanisms developed in the hypermedia field. The investigation takes form of a survey on linking mechanisms supported by earlier hypermedia systems and of the use context these mechanisms were designed to support. A more thorough comparison between XLink and two open hypermedia link formats, OHIF and FOHM, is conducted focusing on transformations between the formats. Based on these comparisons, the implementation of an XLink based prototype (Xspect) is presented. Utilising standard Web technologies, Xspect implements an isomorphic transformation system between OHIF and XLink. The prototype supports navigational hypermedia in the form of links and annotations augmented onto existing Web pages and guided tours presented as interactive maps. Furthermore, the prototype explores the possibilities of visualising advanced link structures using technologies readily available in standard Web browsers. The experiences with the implementation of the Xspect prototype and the comparison of XLink to hypermedia link formats forms a critique and evaluation of XLink as the linking format for the next generation Web and as an interchange format for open hypermedia systems. The evaluation results in the formulation of a "best praxis" on the use of XLink in these contexts.

Yeliz Yesilada, Browsing Tables When You Cannot See Them, University of Manchester, Sep 2000, MSc.
Abstract | Full document

The Internet relies on visual communication and thus causes difficulties to the Visually Impaired. Many Visually Impaired people access the Internet, particularly the World Wide Web (WWW), by listening to synthetic speech generated from machine-readable text. For ordinary text this works well. This is not, however, the case when reading tabular material; tables have many different forms of structure and many different ways of usage, and support a multiplicity of tasks. This project addresses the problems Visually Impaired people have in browsing and reading tables and proposes solutions. We first analyze tables in order to understand the tasks undertaken by Visually Impaired people when browsing and reading tables and we review the recent technological advances in order to determine the capabilities they provide and their inefficiencies. These frameworks form the basis of a system developed to browse and read tables. The aim of the system is to enable Visually Impaired people access to tables as sighted people do and is named as EVITA, which stands for Enabling Visually Impaired Table Accessing. Finally we describe an experiment that was conducted for evaluating the system.