13553 research outputs found
Sort by
Partition First, Detect Faster: A Multi-Partition Subgraph Framework for BGP Anomaly Detection
No full textThe Border Gateway Protocol (BGP) serves as the foundation of inter-domain communication between Autonomous Systems (AS), enabling global Internet connectivity. However, due to the lack of built-in security mechanisms, BGP remains vulnerable to anomalies such as route hijacks and route leaks, which can severely compromise routing stability and availability. Existing anomaly detection approaches often fail to capture the structural complexity of AS-level topologies and overlook computational efficiency, limiting their scalability and real-world applicability.To address these limitations, this thesis introduces a subgraph-based framework designed to achieve high detection accuracy while ensuring computational efficiency in BGP anomaly detection. By partitioning the AS-level topology into structurally cohesive subgraphs using Louvain, Leiden, and -hop strategies, the framework supports localized analysis of routing behaviour while significantly reducing the cost of structural feature extraction. Anomalies are identified by capturing deviations in centrality-based features through an unsupervised learning model, allowing effective detection of inter-domain routing disruptions.We evaluate the proposed framework on real-world BGP anomaly events across diverse scenarios and vantage points. Compared to whole-network baselines, subgraph partitioning significantly reduces the computation time of betweenness and closeness centrality, in some cases by nearly 99\%, while maintaining comparable detection accuracy. For instance, during the CenturyLink incident, closeness computation time at the WIDE collector was reduced by 95.8\%, and betweenness by 98.86\%, without performance degradation. These findings validate the proposed “partition first, detect faster” strategy and demonstrate its potential for scalable, real-time anomaly detection in large-scale inter-domain networks.</p
Making a Movie Camera
No full textThis thesis began without a thesis. It started instead with a curiosity, an internal itch that turned into an exploration of how I see the world and how I try to make things within it. What emerged is a body of work that isn’t so much about answering a question, but about discovering what the question even is. Rather than building on fixed theories or predetermined frameworks, this work let itself unfold through the act of doing. The ideas were allowed to form in their own time, often messy and unexpected. This allowed for a design that was always evolving, just like a conversation. The work led the way. This was never about control, it was about listening. At the centre of this is the idea that design could behave more like dialogue, always shifting and open. That imperfect communication might express something more honest and personal than clean-cut words ever can.This project resists the idea of clarity being a goal. Instead, designing the bits that machines can’t do. This is a practice of design not through knowing, but feeling around in the dark, through embracing the untranslatable. Along the way, it became a kind of antithesis to clear words themselves. This thesis sits between Kentridge’s words of “We need a mistranslation of the world to understand it” (Kentridge, 2022) and Wittgenstein’s widely spread, “The limits of my language are the limits of my world” (Wittgenstein, 2001 , p.68). In a world increasingly defined by clarity, certainty, and machine-like communication, this thesis moves in the opposite direction. It values mistranslation over explanation, feeling over definition. It explores the limits of communication, embracing the idea that sometimes, to understand the world, we need to let it speak in ways we can’t fully understand.</p
Photobleaching of Neutron Radiation Induced Attenuation of Optical Fibers at Liquid Nitrogen Temperature
No full textIn this work, the effect of neutron radiation on optical fibers at cryogenic temperature was investigated and the effect of photobleaching as a way to mitigate the damage caused by this type of radiation was demonstrated. A neutron beam supplying a neutron flux of around 1.5 x 1010 neutrons/cm2 s corresponding to an absorbed dose rate of 0.04 Gy(SiO2)/s was used, which enabled reaching fluences as high as 1015 neutrons/cm2 (equivalent absorbed doses of 2.3 kGy(SiO2) while maintaining the optical fibers at 77 K with a liquid nitrogen bath. A comparison was made between the radiation-induced attenuation (RIA) dynamics of standard (Ge-doped), photosensitive (high Ge-doped), and rad hard (F-doped and pure silica core) fibers. The effectiveness of four power levels (4.5, 1, 0.5, and 0.1 mW) of photobleaching light at 1050 nm at mitigating RIA at 1550 nm was also studied. The recovery of the fibers upon warm-up from 77 K to room temperature was recorded as well as a comparison between RIA of pristine fibers and pre-irradiated fibers
White Phosphorus Activation by Indyl Anions
No full textThe research described in this Thesis covers the synthesis of 8 new xanthene-based indyl anions. This includes the first example of indium anions supported by all group 1 metals (Li-Cs). The differences in their structural forms have been investigated, as well as their reactivity towards the activation of white phosphorus.Chapter One provides a targeted literature review of the group 13 anions, specifically, the aluminyl anion and indyl anion and their reactivity towards small molecule activation and metal-metal bond formation. The Chapter concludes with a targeted review of the low-valent group 13 metal complexes towards white phosphorus activation.Chapter Two extends the family of indyl anions by synthesising various ligand systems with varying degrees of steric bulk. During the synthesis of the first indyl anion in this Thesis, a new method of synthesising the indyl anion via a salt metathesis reaction was discovered, providing access to the low-valent anionic indium complexes without the requirement for harsh reducing agents.Chapter Three investigates the new salt metathesis method to synthesise the series of alkali metal (Li-Cs) supported indyl complexes. However, this method was unsuccessful for the lithium and caesium indyl anions. An alternative metathesis reaction with the lithium and caesium iodides was found to react with the potassium indyl which facilitated access to the lithium and caesium indyl anions.Chapter Four details the isolation of the series of indyl anions with increased stability through the coordination of TMEDA to the group 1 counter ion. Structural comparisons were made between the TMEDA-group 1 (Li-Rb) indyl complexes, before exploring their ability to reduce white phosphorus.Chapter Five provides a comparative study of the first ‘stable’ indyl anion synthesised in this Thesis with unsaturated bonds and the chalcogenides, aiming to compare the isolated products to those of the aluminyl anion. The results in this Chapter further highlight the reactive capabilities of the indyl anion and how it may react as a heavy aluminyl anion analogue.</p
Full scores and lead sheets reductions of transcriptions that accompany <i>The DAWn of a New Fusion: Compositional Styles of Pianist/Producers in the Age of Web 2.0</i>
No full textFull scores and lead sheets reductions of transcriptions that accompany Michael Crawford's Doctorate of Musical Arts (Jazz Composition) exegesis 'The DAWn of a New Fusion: Compositional Styles of Pianist/Producers in the Age of Web 2.0'.</p
'Frequently and Fundamentally Misapprehended': The Scapegoat Rituals of Ancient Greece
No full textEvidence for the performance of scapegoat rituals – wherein a member of the population was abused and expelled for the purpose of city-wide purification – survives for a remarkable number of communities across the ancient Greek world. However, this evidence tends to be both fragmentary and late. The issue is compounded by the characterisation of scapegoat rituals by early scholars of Greek religion as rare, violent, and primitive, a characterisation which has continued to influence more recent scholarship. This has contributed to our poor understanding of the context in which scapegoat rituals were performed and the place of these rituals in the wider scheme of Greek religion. This thesis re-examines the evidence for Greek scapegoat rituals performed in the archaic and classical periods with a specific focus on understanding their relationship to other rituals and myths. Accordingly, I first identify the common features of the – often highly variable – Greek scapegoat rituals through an examination of our evidence. I then explore the performance of scapegoat rituals in the context of the Athenian cults of Apollo and the Thargelia festival. I argue that the regular place of scapegoat rituals within the Athenian festival calendar demonstrates that they were not so esoteric as the scholarship might suggest. Finally, I investigate the relationship between scapegoat rituals and selected elements of Greek myth and cult: the sacrifice of parthenoi, the performance of hero cult, and the myths of scapegoat-like figures. In doing so, I go some way towards contextualising scapegoat rituals within the scheme of Greek religion and, in turn, demonstrating their resonances with other aspects of Greek myth, cult, and society.</p
SUPPORTING THE SENSEMAKING AND ADOPTION OF HUMAN-AI COLLABORATION
No full textOrganizations are undergoing significant changes in the nature of work they perform due to rapid shifts in workplace dynamics and technological advancements. While it is important to acknowledge the role of technology in integrating these advancements into work practices, it is equally essential to holistically understand what other factors, beyond technology, are crucial in facilitating the adoption of these new technologies. Despite the growing interest in Human-AI Collaboration (HAIC) among organizations, many face uncertainty in understanding its potential benefits, risks, obstacles, and other key considerations for integrating it into existing workflows. This interest has been further amplified by the recent introduction of generative AI-powered tools like ChatGPT, which mimic human-like interactions. However, organizations continue to struggle due to a lack of proper insights to identify potential obstacles, evaluate the impact of HAIC, and determine how best to integrate it into their operations.Considering HAIC as a pivotal development in the changing nature of work, this thesis explores and deepens the understanding of HAIC in detail, focusing on its layers of change, sensemaking and adoption. While there is growing interest in adopting Artificial Intelligence (AI), there remains a limited understanding of how organizations approach the change, sensemaking and adoption of HAIC within their workflows. Existing studies predominantly emphasize technological advancements, with limited focus on how humans and AI can collaborate effectively by leveraging their complementary skills in the workplace. Furthermore, although some studies highlight enablers for HAIC, such as technological developments, there is insufficient attention given to a comprehensive overview of obstacles and the organizational changes required for its successful adoption.This study seeks to address these research gaps by conducting empirical research through case study analysis, providing qualitative insights to enhance the understanding of change, sensemaking and adoption of HAIC. It explores and extends the understanding of how organizations approach changes required for HAIC by identifying signs of change in the workplace resulting from recent transformations (adoption of remote work) to examine the socio-technical interplay. Additionally, it investigates how organizations can use HAIC use cases to support prospective sensemaking by anticipating the impact of emerging technologies like AI, enabling them to make informed decisions about AI adoption by understanding its strategic value. To deepen the understanding of HAIC adoption, this study examines the obstacles organizations face in adopting HAIC. To achieve this, this research aims to answer two research questions: How can organizations make sense of Human-AI Collaboration (RQ1) and What are the obstacles to the adoption of Human-AI Collaboration? (RQ2). To address these questions, this thesis comprises four interrelated studies. The first study is a systematic literature review that examines workplace changes driven by the adoption of remote (or hybrid) work, positioning it as a precursor to HAIC. Richter and Richter (2024) reconceptualize hybrid work beyond its traditional definition, which only includes in-person and remote arrangements. They emphasize the importance of considering the impact of HAIC and other virtual technologies on work models. Hence, they argue that hybrid work should also encompass AI integration to better understand how humans and AI interact in both physical and virtual environments. Based on this conceptualization, the signs of change observed in the adoption of remote work serve as both a historical and conceptual precedent, helping to draw parallels between remote work adoption and the anticipated adoption of HAIC. Building on the first study, the second study is an empirical investigation of the layers of change in a case study, exploring the parallel evolution of technological, work practice, and cultural layers within organizations essential for HAIC adoption. The transition to remote work provides a valuable precedent for understanding how organizations navigate technological shifts with the support of social factors. The insights gained from this transformation inform the analysis of the layers of change in HAIC adoption, which represents the next stage in workplace evolution. Focusing on the same case study, the third study delves deeper into how HAIC use cases can support prospective sensemaking within organizations, aiding in a better understanding of HAIC adoption. Since HAIC adoption may be hindered by various obstacles, the fourth study identifies key barriers and provides recommendations to address these challenges.This thesis identifies key findings across several areas. Firstly, it provides insights into potential workplace changes by adopting a socio-technical approach to envision hybrid work. These signs of change in the adoption of remote work serve as a model for understanding HAIC adoption. Based on the insights gained from this transformation, analysis of HAIC considers multiple dimensions of organizational change to thoroughly examine executives’ expectations regarding the reconfiguration of technology, its integration into work practices, and the development of a supportive organizational culture—all of which are critical for the successful adoption of HAIC. Additionally, the study highlights how HAIC use cases can support prospective sensemaking by analyzing these use cases across various functional areas and group tasks. This structured understanding helps organizations identify where and how HAIC can add strategic value while balancing the opportunities and risks associated with its implementation. Secondly, the study identifies key obstacles to HAIC adoption, categorizing them into knowledge, skills, and attitudes to provide a holistic overview. It offers recommendations to overcome these challenges through low-road, high-road, and transformative integrations. This research contributes to both the theory and practice of knowledge related to HAIC by addressing prevailing research gaps and expanding the understanding of its change, sensemaking and adoption. Theoretically, it provides insights by exploring the sensemaking of HAIC, identifying the change dimensions required for its adoption and their interplay, presenting a comprehensive overview of obstacles to adoption. Practically, it offers several key contributions. First, it presents an overview of the technological, work practice, and cultural changes necessary for HAIC adoption. Second, it provides a structured matrix that helps organizations understand how HAIC can be applied across various business functions and collaborative tasks, supporting their sensemaking and strategic AI implementation. Third, it introduces a framework for organizations to identify obstacles and derive recommendations for effectively adopting HAIC. The findings of this study are significant for various stakeholders, including researchers, practitioners, and policymakers, as they advance the understanding of HAIC while helping organizations navigate the complexities of HAIC in modern workplaces. This study not only deepens the understanding of sensemaking and adoption of HAIC but also bridges the gap between theory and practice by supporting the integration of AI into organizational workflows, facilitating effective human-AI collaboration.</p
"Experiments in Life": Narrating Relations in George Eliot and Thomas Hardy
No full textThe realist novelists of the mid-late nineteenth century inherited their predecessors’ conviction that the individual protagonist was the most fitting vessel for figuring the larger forces (natural, social, historical) which shape our experience of the world. But their mimetic project was complicated by developments in the scale and methods of scientific inquiry. As investigators increasingly supplemented classical empirical methods with more abstract theories and statistical approaches, the individual case or instance—the familiar territory of realist representation—was destabilised as a site for generating knowledge about the world. How did realist novelists negotiate between their accustomed scope and a scope which comprehended new understandings of a “massively distributed reality” (Rosenberg) beyond the reach of the senses?This thesis provides two answers, reading two works by novelists writing at the frontiers of British realism: George Eliot’s The Mill on the Floss (1860) and Thomas Hardy’s The Return of the Native (1878). My study is framed by Eliot’s notion of her novels as “experiments in life”: attempts to trace the relations between “individual experience” and the “formula” or law which is supposed to govern that experience. If both novels are committed to the individual scale, they also figure some of the formal and conceptual challenges which emerge from its inability to be neatly integrated into broader structures for understanding reality. My readings are informed by, but not dependent on, notions of a dialogue between Victorian literature and science. Rather than tracing these connections, this thesis gives an account of how these literary experiments with narrative meet or respond to the challenges posed by their intellectual context.Chapter 1 intervenes in critical readings that interpret The Mill alongside nineteenth-century geology. It argues that the novel’s flood, interpreted within the novel’s uneven plot structure, formalises a key scalar challenge posed by Charles Lyell’s uniformitarianism: how to negotiate between minute observations (which furnish the basis of historical knowledge) and the uniformity which only emerges at “a long average” (Whewell). Where Sally Shuttleworth influentially argued that the novel is riven by a conflict between incommensurable “models” of geologic history, I draw on revisionist histories of science to interpret the flood as a collision between historical scopes, problematically embodied by the narrator’s suspension between the impulses of memory and history.Chapter 2 turns from the individual in time to the individual in space, uniting two impulses in Hardy criticism by attending both to the minutiae of individual encounters between “Character and Environment” and the social and geographical structures that mediate those relationships. In The Return, the “vast sum of conditions” (Eliot) that inflect the individual’s perception of Egdon Heath are folded into focalised descriptions, vexing the boundary between observer and observed. This entanglement is also figured by the novel’s interest in traffic across borders, its complex positioning of its characters across categories like ‘native’ and ‘alien’, and by passages which historicise Egdon Heath as an “unenclosed … wild” in an increasingly enclosed nation. By undermining the binaries it establishes, the novel highlights the open and reciprocal connections between the discrete categories we invent to make sense of a world of essentially continuous relations.</p
The Ocean at the Bottom of the Garden
No full textOceans are good for cities, but historically cities have been mostly bad for oceans. Oceans feed cities, create pleasurable environments for citizens and hide or disperse waste readily (to a point). But the progressive deterioration in oceanic environmental quality that we have experienced with the evolution and expansion of urbanisation has to date spawned few ways of improving, cultivating and repairing the urban/marine relationship. This could be in part because the framework through which cities seek to respond has been mostly centred around engineering and the sciences, whereby oceans are either examined as pristine environments to be protected, or conversely as a threat from which cities need themselves to be guarded.For coastal cities, the urgency of climate change combined with the cultural, physical and economic investments which are the means of their existence, the problem of marine landscapes is a pressing one.What are we doing to identify new futures for the urban marine landscape through practices of design? In the field of landscape architecture we could start by asking whether the marine environment a ‘landscape’ able to be shaped, and if what conventions we might carry forward or need to develop in order to engage with it.</p
Evaluating the mechanism of action of a novel heparan sulfate mimetic in an animal model of multiple sclerosis
No full textMultiple sclerosis is a neuroinflammatory condition in which cells of the immune system invade and attack the central nervous system. This results in widespread inflammation, demyelination, and neurodegeneration leading to a host of neurological symptoms. People with multiple sclerosis commonly experience relapses in symptoms due to new waves of immune cells entering the central nervous system. As such, treatment is aimed at global immune suppression or inhibition of the migration of immune cells into the central nervous system. In healthy individuals, the brain and spinal cord are normally protected from the immune system by the blood brain barrier. This highly organised structure utilises many mechanisms to prevent unrestricted access to the central nervous system by circulating immune cells. One specialisation of the blood brain barrier is the double basement membrane formed by central nervous system endothelial cells and the glia limitans of astrocytic endfeet. This dense glycocalyx provides both a physical barrier to immune cell entry and is critical in maintaining blood brain barrier organisation through sequestered of trophic factors and signalling molecules. The final step in immune cell diapedesis across the blood brain barrier is the enzymatic breakdown of basement membrane components such as heparan sulfate proteoglycans. One such enzyme is heparanase, which is released from activated antigen specific T cells in animal models of multiple sclerosis. By cleaving heparan sulfate proteoglycans, heparanase facilitates the migration of autoreactive immune cells into the central nervous system. Therefore, inhibition of heparanase is an attractive treatment strategy for multiple sclerosis. Heparanase can be inhibited by small heparan sulfate-mimicking, sugar-based compounds. Multiple iterations of heparan sulfate-mimetics have been developed over the years but so far most of these have not progressed to clinical trials due to high anticoagulant effects and complexity in synthesis. Tet-29 is a novel dendrimer-based heparan sulfate-mimetic that exhibits high heparanase inhibiting abilities without the associated anticoagulant effects. This thesis investigated the disease modifying capability of Tet-29 in experimental autoimmune encephalomyelitis, a mouse model of multiple sclerosis. Treatment with Tet-29 both before and after established disease ameliorated physical symptoms of paralysis. This disease modifying effect was due to a significant reduction in infiltration of T lymphocytes into the central nervous system of Tet-29 treated mice. Further investigation found that this reduction in central nervous system trafficking was due to a rescue of disease-induced blood brain barrier permeability by Tet-29, rather than peripheral immune depletion. Moreover, Tet-29 did not disrupt homeostatic trafficking into the central nervous system, suggesting it does not interfere with surveillance. Disrupted immune surveillance of the central nervous system is a major drawback of other multiple sclerosis treatments on the market. The collective findings of this thesis provide evidence that Tet-29 could have a promising therapeutic benefit in multiple sclerosis. By targeting blood brain barrier breakdown rather peripheral immune responses, Tet-29 exhibits not only a novel mechanism of action, but also potentially avoids many of abhorrent side effects associated with immunosuppressant therapies. The work presented in this thesis presents clear evidence that further preclinical investigation into the suitability of Tet-29 as a therapeutic agent for multiple sclerosis is warranted.</p
Access Repository Dashboard
Do you manage Victoria University of Wellington? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!