85000 research outputs found
Sort by
Restoring Singapore's Food Collective Actualisation Through Regenerative Agrarian Dynamics
No full textThis practice-led research investigates the possibilities and constraints of restoring food collective actualisation in Singapore through regenerative agrarian dynamics, situated within a land-scarce, hyper-urbanised island-state shaped by neoliberal governance. Drawing on a decade of embedded practice with Edible Garden City (EGC), a social enterprise pioneering community-centric urban farming—this thesis explores how urban food sovereignty can be reinterpreted beyond its conventional agrarian roots and realised through tactical spatial interventions, community activation, and care-based farming models. Through the enactment of over 280 edible gardens and farming initiatives, the research examines the evolving role of urban agriculture as more than a technical solution to food production, but as a socio-ecological catalyst for intergenerational engagement, community cohesion, and inclusive urban space-making. The inquiry contextualises Singapore’s food security agenda against broader geopolitical and ecological shifts in the Anthropocene and proposes new evaluative frameworks such as the Urban Food Sovereignty Index and the Landscape Nutrition Index, to measure nutrient equity and participatory agency in urbanised contexts. This thesis contributes new knowledge through reflective documentation of achievements, failures, frictions, and adaptations within the practice of EGC, offering insights into the sociopolitical dynamics of urban agrarian reform. While acknowledging the limitations of land availability and systemic constraints, it foregrounds the critical role of relational, place-based, and care-oriented farming practices in shaping regenerative food futures for cities.</p
Shaping the future of education: a cluster analysis of generative AI’s transformative impact
No full textPurpose This study aims to investigate the transformative impact of Generative Artificial Intelligence (Gen-AI), particularly ChatGPT, on education. Through comprehensive bibliometric and content analysis, this study maps publication trends, identifies key research themes, uncovers gaps in the literature and explores future directions for effectively integrating AI into educational systems. Design/methodology/approach A systematic review of 817 articles published between 2021 and 2024 was conducted to explore the evolving landscape of Gen-AI in education. Using bibliometric and content analysis, this study used coword analysis, thematic mapping, cluster analysis and bibliometric coupling to identify trends, gaps and the structural and conceptual frameworks underlying the integration of Gen-AI in educational settings. Findings The analysis identified four key thematic clusters: Gen-AI as a driver of educational transformation, its impact on curriculum and pedagogy, ethical and integrity considerations in higher education and its role in enhancing creativity and knowledge. This study proposes a conceptual framework with 10 propositions and 12 research inquiries, emphasizing personalized learning approaches and robust ethical safeguards. Practical implicationsThis research offers actionable insights for educators, policymakers and AI developers, providing a roadmap for the responsible integration of Gen-AI into educational strategies. It emphasizes fostering innovation while addressing concerns about ethical, social and academic integrity.Originality/valueThis study offers a comprehensive synthesis of the rapidly expanding research on Gen-AI in education, with a particular emphasis on ChatGPT. Analyzing 817 peer-reviewed articles from 2021 to 2024, this study combines bibliometric, thematic and content analyses to identify four research clusters and track their evolution. Aligned with UNESCO’s AI Ethics Recommendation, it provides a conceptual framework, 10 theoretical propositions and 12 research questions, delivering practical insights and a strategic agenda for researchers, educators and policymakers.</p
Surface Functionalised Liquid Metals for Catalysis and Green Fuel Synthesis
No full textLiquid metals (LMs) and their alloys, with their inherent fluidity and unique electrohydrodynamic properties, have emerged as promising catalysts for energy conversion and chemical synthesis. Unlike traditional solid catalysts, LMs exhibit self-healing surfaces, tuneable surface tension, and the ability to dissolve metals in low oxidation states, reducing catalyst deactivation and enhancing reaction efficiency. However, their high surface tension poses challenges in scaling catalytic reactions and optimizing interfacial interactions.
This thesis investigates new avenues to alter the surface characteristics of LM and utilize their catalytic activity. To begin with, systematic designs of LM-based electrocatalysis electrodes are investigated with a focus on LM-electrolyte maximization for improved catalytic activity. Thus, a scalable system for the creation of microdroplets of liquid metals is established through electrohydrodynamic modulation. It is a method involving the use of electric potentials, enabling precise control over the size of liquid metal droplets while keeping the energy consumption low. The examination of oxide layer modulation reveals its significant influence on the tuning of LM surface tension, particularly in relation to droplet formation and stability.
Moreover, the catalytic potential of LM micro- and nano-droplets is examined, with an emphasis on improving the efficiency of the oxygen reduction reaction (ORR). Additionally, this work introduces an on-demand hydrogen production system, where trace amounts of platinum activate gallium-based alloys to facilitate rapid hydrogen evolution. This methodology provides a sustainable and energy-conserving pathway for hydrogen production while simultaneously producing 2D nanoporous platinum electrocatalysts characterized by exceptional ORR activity.
This thesis enhances the comprehension of LM catalysis through the integration of experimental, theoretical, and computational insights, offering scalable solutions for electrocatalytic applications. The results of this research significantly advance the overarching progress in sustainable energy technologies and liquid metal-based reaction systems.</p
Postbiotics: A Promising Approach to Combat Age-Related Diseases
No full textDietary patterns have been identified as one of the most important modifiable risk factors for several non-communicable diseases, inextricably linked to the health span of older people. Poor dietary choices may act as triggers for immune responses such as aggravated inflammatory reactions and oxidative stress contributing to the pathophysiology of several ageing hallmarks. Novel dietary interventions are being explored to restore gut microbiota balance and promote overall health in ageing populations. Probiotics and, most recently, postbiotics, which are products of probiotic fermentation, have been reported to modulate different signalling biomolecules involved in immunity, metabolism, inflammation, and oxidation pathways. This review presents evidence-based literature on the effects of postbiotics in promoting healthy ageing and mitigating various age-related diseases. The development of postbiotic-based therapeutics and diet-based interventions within a personalised microbiota-targeted approach is proposed as a possible direction for improving health in the elderly population. Despite growing evidence, the data regarding their exact mechanistic pathways for antioxidant and immunomodulating activities remain largely unexplored. Expanding our understanding of the mechanistic and chemical determinants of postbiotics could contribute to disease management approaches, as well as the development of and optimisation of biotherapeutics.</p
A Conceptual Framework for Updating Urban Infrastructure in City Digital Twins
No full textThis work presents a conceptual framework for updating urban infrastructure by separating the update process into two loops, the data loop and the application loop. The key aspect of the data loop is intelligent change detection followed by validation strategies to maintain an up-to-date catalogue of city infrastructure. The updated catalogue can subsequently utilised to create user-centric digital twins. The advantages of using such a framework are presented, in addition to discussing the challenges and their possible solutions. Further, the scope of integrating advanced change detection approaches, new data structures such as Knowledge Graphs and Voxels, and AI methods such as the recent Large Language Models (LLM) and Continual Learning are discussed.</p
In The Wake of Light
No full textBackgroundIn the Wake of Light is part of a larger funded project that films Ho Chi Minh City through a series of motorbike journeys. This creative practice research output offers a sensory counter-mapping of the space between the motorbike and the surrounding urban environment. It aligns with sensory and embodied studies, while also contributing to the understanding of intangible heritage in Ho Chi Minh City. The work follows in the footsteps of Ed Ruscha and his book Every Building on the Sunset Strip, which presents a continuous photographic sequence of Sunset Strip in the USA.ContributionThe film documents a journey through District 1, the central district of Ho Chi Minh City, filmed at a 90° angle to the direction of travel. This re-centering of the motorbike driver’s and passenger’s peripheral vision emphasizes the embodied experience of passing events and their associated atmospheres. The use of blur and colour tinting intensifies the reading of the film sequences. These disorienting visual treatments heighten the viewer’s awareness of what is passing across the screen. This close observation of everyday experiences of motorbike travel contributes to a deeper understanding of the practice as an element of intangible heritage. The film offers a layered experience that entangles not only the act of travelling by motorbike but also the activities and events that unfold in the city streets.SignificanceIn the Wake of Light was selected by a review panel for exhibition at iNVENTX, held at the Faculty of Creative Multimedia, Multimedia University in Cyberjaya, Malaysia. At this exhibition, the work received a Gold Award. iNVENTX, themed around “Sustainaissance,” is an initiative that invites artists and designers from the region to exhibit their work in a gallery setting. The exhibition catalogue has been issued with an ISBN - e ISBN 978-629-7709-18-5</p
Supercomputing Multi-Ligand Modeling, Simulation, Wavelet Analysis and Surface Plasmon Resonance to Develop Novel Combination Drugs: A Case Study of Arbidol and Baicalein Against Main Protease of SARS-CoV-2
No full textBackground/Objectives: Combination therapies using traditional Chinese medicine and Western drugs have gained attention for their enhanced therapeutic effects and reduced side effects. Toujie Quwen Granules (TQG), known for its antiviral properties, particularly against respiratory viruses, could offer new treatment strategies when combined with antiviral drugs like arbidol, especially for diseases such as Coronavirus disease. This study investigates the synergistic mechanisms between arbidol and components from TQG against the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) main protease (Mpro). Methods: We identified compounds from TQG via existing data. Multi-ligand molecular docking, pharmacokinetic/toxicity screening, and preliminary simulations were performed to assess potential synergistic compounds with arbidol. UPLC-Q-Exactive Orbitrap-MS verified the presence of these compounds. Extended simulations and in vitro assays, including Luciferase and surface plasmon resonance, validated the findings. Results: Five compounds interacted with arbidol in synergy based on docking and preliminary dynamics simulation results. Only Baicalein (HQA004) could be identified in the herbal remedy by untargeted metabolomics, with ideal pharmacokinetic properties, and as a non-toxic compound. Extended simulations revealed that HQA004 enhanced arbidol’s antiviral activity via a “Far” Addition Mechanism #2, with an optimal 2:1 arbidol:HQA004 ratio. The movements of arbidol (diffusion and intramolecular conformational shifts) in the system were significantly reduced by HQA004, which may be the main reason for the synergism that occurred. In vitro experiments confirmed an increased inhibition of Mpro by the combination. Conclusions: HQA004 demonstrated synergistic potential with arbidol in inhibiting Mpro. The development of combination therapies integrating Western and herbal medicine is supported by these findings for effective antiviral treatments.</p
‘It’s what we’ve always done’: the netball dress, tradition and white femininity in Australia
No full textNetball, a sport historically played and controlled by women in Australia, has long been identifiable by the iconic, albeit evolving, uniform of the netball dress. However, in 2022, amidst a broader shift in the conversation around women’s uniforms, Netball Australia announced new, more expansive uniform guidelines. In this context I examine the historical and contemporary tensions of the netball dress in Australia. As a feminine-coded activity with roots in British imperialism, netball and the netball dress offer insights into the implicit and explicit expectations of (racialized) femininity imbued in active women’s uniforms and clothing (Horton et al. 2016; Marfell 2019). Adopting a critical whiteness and intersectional feminist lens, and drawing on interviews with netballers, I consider the significance and impacts of the netball dress in shaping and upholding a particular version of white femininity in the Australian netball context.</p
Mood as the Mediator of the Relationship Between Interoceptive Sensibility and Alexithymia
No full textAlexithymia is a personality construct characterised by difficulties describing and identifying
emotions. Alexithymia was evident to be associated with interoception, the ability to perceive
and interpret internal bodily signals. There is a limited investigation on self-evaluated
interoceptive sensibility aspect (IS) and its link with alexithymia and covariates. Therefore,
the present cross-sectional design established the relationship between alexithymia and IS,
assessed by the Multidimensional Assessment of Interoceptive Awareness (W. E. Mehling et
al., 2012) and the Body Mindfulness Questionnaire (Burg et al., 2017) (N = 161). The effects
of potential covariates were also examined. Our study reported the significant inverse
correlation between various aspects of alexithymia and IS. Especially, based on regression
models, we proposed and scrutinised “experiencing body awareness” and “trusting body
awareness” as fundamental factors of IS in relation to alexithymia. Crucially, the present
research claims the mediating effects of depression and anxiety on this relationship. These
findings provided the new pathway to understand the interaction between IS, mood and
alexithymia, thus shed light on the influence of mindful attention style and trusting attitude in
IS as well as the alexithymia subtypes.</p
Secure and Robust Collaborative Learning for Internet-of-Things Data Analytics
No full textThe rapid growth of the Internet of Things (IoT) and Artificial Intelligence (AI) has transformed industries by enabling efficient data analytics and intelligent decision-making. Among AI techniques, deep learning models often require large datasets for effective training. However, this poses challenges in privacy-sensitive domains such as healthcare and government due to strict data-sharing regulations. Limited access to high-quality data reduces the efficacy and generalization of the AI model, making collaborative learning techniques such as Federated Learning (FL) and Split Learning (SL) essential. These approaches enable multiple entities to train models without sharing raw data, preserving privacy while improving AI model performance. As a result, collaborative learning has gained popularity across industries, but significant challenges remain. To begin with, privacy concerns persist even with the use of collaborative techniques. Model parameters, although they replace direct data sharing, can still reveal insights into the underlying data, posing risks of data reconstruction or information extraction by adversaries. To mitigate these risks, we propose a differential privacy-based, privacy-preserving collaborative learning model for resource-constrained IoT devices. The proposed method involves perturbing the locally trained model to prevent adversaries from inferring the data used during training.Moreover, ensuring data integrity during exchange processes in collaborative learning is crucial. Data, including model parameters, are vulnerable to alterations during transmission over networks, which could compromise the accuracy and reliability of collaborative learning outcomes. To address this, we developed a blockchain-based, privacy-preserving IoT data analysis model for heterogeneous collaborative learning. In this work, blockchain technology is introduced to mitigate data tampering during the exchange of trained models. Additionally, a model cross-validation process among participants is implemented to ensure the equitable contribution of each participant's model. Blockchain technology is also used to protect the integrity of cross-validation results.In addition, the presence of resource heterogeneity among edge devices in IoT systems introduces complexities in collaborative learning. Variations in computing resources and network bandwidth between devices may cause delays or even prevent model convergence, thereby affecting overall learning efficiency. In response to this challenge, we propose weight-based asynchronous model aggregation techniques for collaborative learning. These techniques enable more efficient learning by ensuring that stalled devices do not hinder fast model convergence, leading to improved model performance.Finally, data heterogeneity poses significant challenges to collaborative learning frameworks, particularly in IoT data analytics scenarios. Imbalances in data distribution, where certain entities possess disproportionate amounts of specific data types, can skew model predictions and hinder generalization capabilities. To address this challenge, we introduce clustering-based collaborative learning techniques tailored for non-independent and identically distributed (non-IID) data. We redefine the objective function in collaborative learning for non-IID scenarios, focusing on minimizing data disparity among participants. By clustering clients according to data statistics, we aim to optimize this new objective function. All in all, this thesis examines collaborative learning through four key challenges: privacy preservation, data integrity, resource heterogeneity, and data imbalance. To address these challenges, we propose algorithms leveraging privacy preservation techniques, blockchain-based validation, asynchronous AI model aggregation, and clustering techniques. We believe that this research represents a significant step towards building a more secure, robust, and efficient collaborative learning framework in decentralized and resource-constrained settings.</p