48651 research outputs found
Sort by
From Metrics to Meaning: Time to Rethink Evaluation in Human–AI Collaborative Design
No full textAs AI systems increasingly shape decision making in creative design contexts, understanding how humans engage with these tools has become a critical challenge for interactive intelligent systems research. This paper contributes a challenge to rethink how to evaluate human--AI collaborative systems, advocating for a more nuanced and multidimensional approach. Findings from one of the largest field studies to date (n = 808) of a human--AI co-creative system, The Genetic Car Designer, complemented by a controlled lab study (n = 12) are presented. The system is based on an interactive evolutionary algorithm where participants were tasked with designing a simple two dimensional representation of a car. Participants were exposed to galleries of design suggestions generated by an intelligent system, MAP--Elites, and a random control. Results indicate that exposure to galleries generated by MAP--Elites significantly enhanced both cognitive and behavioural engagement, leading to higher-quality design outcomes. Crucially for the wider community, the analysis reveals that conventional evaluation methods, which often focus on solely behavioural and design quality metrics, fail to capture the full spectrum of user engagement. By considering the human--AI design process as a changing emotional, behavioural and cognitive state of the designer, we propose evaluating human--AI systems holistically and considering intelligent systems as a core part of the user experience---not simply a back end tool
Assessing the combined effect of gamma radiation and sulfate-reducing bacteria on copper corrosion for deep nuclear waste storage
Full text linkEnsuring the integrity of barriers in geological disposal facilities (GDFs) is crucial for the long term storage of radioactive waste. Copper is considered as a promising canister material due to its corrosion resistance. This study examines the combined effects of external gamma radiation (14 and 28 kGy) and sulfate-reducing bacteria (SRB) on copper corrosion in highly compacted FEBEX bentonite. Results showed that gamma radiation significantly reduces SRB viability, suggesting that these bacteria are likely to remain inactive during the early centuries of GDF operation, when radiation is at its highest level. Microscopic and spectroscopic analyses identified copper oxides, particularly CuO, as the main corrosion products. Gamma radiation was found to delay microbial influenced corrosion by altering the microbial community structure and promoting salt precipitation, including copper sulfates. SRB facilitated the formation of biogenic copper sulfides in unirradiated or those minimally affected by radiation. These findings provide valuable insights into the role of SRB in copper corrosion, broadening our understanding of long-term GDF safety
Antibiotic-Mediated Microbiota Depletion of <i>Aedes aegypti</i> Gut Bacteria Modulates Susceptibility to Entomopathogenic Fungal Infection and Modifies Developmental Factors
Full text linkEntomopathogenic fungi are promising alternatives to synthetic insecticides for the control of vector species, notably the arbovirus vector, Aedes aegypti. The influence of intrinsic mosquito midgut microbiota on host susceptibility to fungal infection and subsequent physiological processes remains poorly understood. Here we treated female Ae. aegypti with the broad-spectrum antibiotic carbenicillin to reduce gut bacterial populations, then exposed them to Metarhizium anisopliae conidia. Female Ae. aegypti offered carbenicillin and then sprayed with fungi had significantly lower survival rates (38.9% ± 1.15) compared to non-antibiotic-treated mosquitoes sprayed with fungus (68.9% ± 0.58). To monitor the kinetics of microbial community recovery, mosquitoes were challenged with conidia at 0, 3, 6, and 9 days following antibiotic removal from the diet. Reduced survival persisted through the 6-day period (survival rates 37.8% to 45.6%), with a significant increase in survival observed 9 days post-antibiotic removal (58.9% vs. control 63.3%), which coincided with recovery of gut bacterial populations. Additionally, antibiotic and fungal treatments reduced egg production, larval eclosion, and pupal formation. These results demonstrate that gut bacteria contribute to mosquito defense against fungal pathogens and support normal reproductive and developmental functions. Understanding the interplay between gut microbiota and entomopathogenic fungi may enhance biological control approaches
Benefits of public engagement in research and barriers to participation: a UK‐based survey of academic scientists and support staff including international respondents
Full text linkPublic engagement is increasingly central to research, especially in biomedical fields, fostering dialogue between scientists and society, building trust and ensuring real‐world relevance. However, as scientific and clinical progress accelerates, the gap between researchers and the public continues to widen, underscoring the need for deeper, more meaningful engagement. Despite the acknowledged value of public engagement for both researchers and the public, we know relatively little about academics' views on opportunities and potential barriers to participation. Using questionnaires and interviews, this study captured insights from 99 researchers and professionals across academic disciplines, career stages and geographical and cultural contexts. Respondents consistently regarded public engagement as an important and rewarding aspect of research, teaching and institutional responsibilities, with the potential to enhance public understanding, acceptance and societal impact. However, enthusiasm was tempered by persistent barriers, including academic workloads, inadequate resources and support, and a lack of formal recognition within career progression. Respondents emphasized the need for systemic reforms to enable greater participation, including tailored training, sustained funding and institutional frameworks that acknowledge and reward engagement. Overall, the findings demonstrate that while motivation for public engagement is widespread, structural and systemic challenges limit its full potential. Addressing these barriers requires coordinated action from universities, funders and policymakers to establish and embed public engagement more consistently as an integral component of academic research and higher education
Ring‐Width Dendrochronology, Isotopic Dendrochronology and Radiocarbon Dating of Timbers From the Spire Scaffold of Salisbury Cathedral, Wiltshire, England
Full text linkTen timbers from the spire scaffold of Salisbury Cathedral were dated using a combination of ring‐width dendrochronology, stable oxygen isotopic dendrochronology and radiocarbon dating. Seven timbers were coeval and assigned a combined empirical felling date range of 1352–1378, which was further refined to 1351–1359 (OxCal 95.4%). These results would indicate that the scaffold was not a 1320s construction but instead built later in the 14th century. The remaining sampled timbers produced a precise felling date of spring 1737 and are coincident with documented repair work in 1738
Blockchain and federated Q-learning-based secure, fault tolerant, and energy efficient framework for ad hoc networks
No full textAn ad hoc network plays a critical role in enabling communication in environments where deploying fixed infrastructure is impractical or infeasible. However, their dynamic topology and decentralized nature make them highly susceptible to failures and security threats. This paper proposes a robust and intelligent framework that addresses these challenges by integrating secure communication, fault tolerance, and energy efficiency. The proposed model makes use of blockchain technology to encourage trust between nodes, works with several nodes in the network to spot their unusual behavior, and uses federated Q-learning for adaptive threat response. The key components of the proposed framework, i.e., identity validation, trust scoring, distributed anomaly detection, and autonomous role management, make the system stable, robust, and energy-efficient. Simulation results of 500 nodes in a dynamic network show that the proposed model provides better performance in packet delivery, fewer false detections, and shorter recovery time in comparison to other systems. Furthermore, the proposed system holds significant promise for critical applications such as battlefield communication, disaster recovery, and remote monitoring, where reliable and secure networking is essential. The novelty of the work is the combination of a lightweight blockchain (MicroChain), Adaptive Cryptographic Engine (ACE), and federated Q-learning into one framework of ad hoc networks. The proposed framework provides high-security, effective resource usage, and responsiveness to the current network environment unlike earlier solutions, which focus on security, power usage, or fault tolerance separately
Working against the backdrop of extreme marginalisation: stigma and the social relational model for the setting of mental health conditions
Full text linkThis paper investigates how employees with mental health conditions (MHCs) experience and respond to working in the contemporary UK workplace. Employing the Social Relational Model (SRM) of disability, the paper positions stigma as an organising structural force that actively produces Social and Relational disabling barriers – impairment effects, barriers to doing, and barriers to being – that shape the working lives of employees with MHCs. Qualitative data from 42 interviewees working for varied employers – including small, medium and large enterprises, public and private sector – reveals how workplace processes and practices assume norms of the ‘ideal worker’, a worker characterised by uninterrupted productivity and emotional stability. We explore how these norms contribute to the stigmatisation of workers managing MHCs and how consequently these workers avoid workplace stigmatisation. By explicitly linking understandings of structural stigma to the SRM, we advance understanding of how stigma operates in often indirect and subtle ways to disable employees with MHCs. Conclusions with implications for HRM include the need to confront normative ideals and institutional practices that sustain stigma by advocating for practices that dismantle stigma, challenge ableist constructs, support diverse mental health experiences and, focus on creating ideal workplaces, rather than continuing to valorise the ideal worker
Normality and Disability in Para Sport and Their Application in the International Paralympic Committee Handbook: A Philosophical Critique
No full textA Hybrid Ensemble Approach for Early‐Stage Diabetes Detection
Full text linkDiabetes has become a critical global health concern, particularly in regions where access to diagnostic facilities is limited. In this work, we propose a hybrid framework that combines extreme gradient boosting (XGBoost) and deep neural networks (DNNs) for early-stage diabetes detection, using soft voting to generate the final ensemble predictions. The proposed framework was evaluated on two datasets: the widely used Diabetes UCI dataset and a newly collected dataset from Nepal. The ensemble method achieved 99% accuracy (ACC) with an area under the curve (AUC) of 1.00 on the Diabetes UCI dataset, and 91% ACC with a 0.96 AUC on the Nepal diabetes dataset, demonstrating its strong generalisability across distinct populations. Compared to individual models, the hybrid approach offered increased stability and a lower rate of false negatives, which is particularly important in clinical contexts. These findings highlight the potential of hybrid machine learning–deep learning models as cost-effective, scalable and generalisable decision-support tools for diabetes risk assessment
Data sources:
Access Repository Dashboard
Do you manage Swansea University Cronfa? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!