30793 research outputs found
Sort by
A novel approach to automating context-driven alternative text generation through purposeful games
Full text linkThis thesis was submitted for the award of Doctor of Philosophy and was awarded by Brunel University LondonAccessibility of the Web is a pervasive issue, owing to the persistence of accessibility barriers (e.g., poor navigation, lack of/unsuitable alternative text (alt text), complex web forms), with significant impact on users with disabilities. Alt text barriers in particular, are some of the most prevalent web accessibility barriers affecting a wide range of media and are underpinned by a lack of understanding and guidelines on what constitutes suitable alt text. Past work has shown that the ‘context in which the image is used in’ is ironclad for suitability yet loosely defined. Whilst there is a need to automate alt text generation, current solutions disregard context in alt text and are lacklustre with regard to suitability. In this research, an empirical exploratory study that investigates the views of web content creators and visually impaired users on suitability is conducted to bridge the functional gap between experiences and best available practice. The first definition of ‘Alt Text Context’ is proposed providing a systematic way to assess when alt text is necessary and what it should convey. Further, the first crowdsourcing game for context-driven alt text authorship and evaluation—TagALTlong—is presented. TagALTlong’s design is informed by relevant literature, empirical qualitative insights and the proposed definition of context in alt text. Following an empirical user study, 125 non-expert players were recruited to play TagALTlong over a six-week period, resulting in 1208 authored and 1836 rated alt text descriptions, respectively. The resulting dataset was used to fine-tune and train an AI model for automated alt text generation to assess whether average human-level alt text quality can be approximated whilst automating the process. Results indicated the improved performance of the model that was fine-tuned and trained on the GWAP-generated dataset compared to pure image processing, subsequently demonstrating the value of the dataset
Energy allocation explains how protozoan phenotypic traits adapt to temperature and nutrient supply
No full textA preprint version is available online at bioRxiv, https://www.biorxiv.org/content/10.1101/2023.07.25.550219v6.abstract. It has not been certified by peer review....This work was supported by the Royal Society Research Grant RG170282 to Andrea Perna
From ‘Modern Slavery’ to Modern Complicity: The corporatisation of western anti-slavery INGOs
No full text........
Pre-Ictal EEG Augmentation Based CDCGAN Model for Epileptic Seizure Prediction
Full text linkData Availability Statement:
The CHB-MIT Scalp EEG Database used in this study is publicly available at https://physionet.org/content/chbmit/1.0.0/ (accessed on 13 December 2025).Epilepsy is a common neurological disorder affecting over 50 million people worldwide, characterised by recurrent seizures accompanied by abnormal neuronal electrical activity. Electroencephalogram (EEG) is a technique for recording brain electrical signals, widely employed for epileptic seizure (ES) prediction due to its high temporal resolution, portability, and cost-effectiveness. However, reliable ES prediction based on EEG remains challenging, primarily owing to the limited duration of recorded pre-ictal states in publicly available datasets and the typically low signal-to-noise ratio (SNR) in non-invasive recordings. To mitigate these issues, we propose a Conditional Deep Convolutional Generative Adversarial Network (CDCGAN), which combines the representational power of Deep Convolutional Generative Adversarial Network (DCGAN) with the categorical conditioning mechanism of Conditional Generative Adversarial Network (CGAN) to generate class-specific EEG samples. By synthesising target samples, CDCGAN aims to alleviate class imbalance and enhance the quality of low-resolution spectral representations. To evaluate the practical utility of generated data, we trained a Convolutional Neural Network (CNN) on the augmented dataset and compared its performance against prior studies. Under the Leave-One-Seizure-Out cross-validation (LOSO-CV) protocol, our method achieved an average AUC of 0.876 at a 60% augmentation rate with 50 training epochs. The AUC improvement relative to corresponding control settings demonstrates that GAN-based data augmentation provides additional effective training samples for ES prediction while preserving task-relevant and discriminative pre-ictal EEG features.This research was funded by the Royal Society (IEC\NSFC\223285) and National Natural Science Foundation of China (General Program) No. 62171073
Combined multi‐metric assessment of diaphragm contractile function in healthy humans: Feasibility, validity and reliability
Full text linkData Availability Statement:
The data that support the findings of this study are shown in the figures, tables and supporting information.The combined use of subcostal ultrasonography and respiratory manometry represents a novel, integrative method for quantifying diaphragm contractile function (force, velocity and power). We evaluated the technical feasibility, construct validity and within-day test–retest reliability of this method during non-volitional, volitional and reflexive respiratory perturbations in healthy adults. Two independent cohorts were studied. In Experiment 1 (n = 10), diaphragm excursion (subcostal ultrasonography) and transdiaphragmatic pressure (Pdi, manometry) were measured during unilateral magnetic phrenic nerve stimulation (non-potentiated and potentiated twitches, paired stimuli at 10–100 Hz) and maximal sniffs. In Experiment 2 (n = 8), the same measurements were obtained during progressive CO2 rebreathing. All protocols were repeated after 20 min of rest. Diaphragm velocity and power were calculated as excursion/time and Pdi × velocity, respectively. Ultrasound analysis was successful in >95% of cases. Potentiated twitches elicited greater Pdi, excursion and power than non-potentiated twitches, with responses increasing at higher stimulation frequencies. Reliability improved with potentiation and high-frequency stimulation and was moderate to excellent for peak responses during sniffs and CO2 rebreathing (ICC3,k = 0.70–0.94) but poor for slope-based measures (ICC3,k ≤ 0.20). During CO2 rebreathing, excursion and velocity correlated strongly with inspiratory tidal volume (r = 0.83, P < 0.001) and mean inspiratory flow (r = 0.69, P < 0.001), respectively. These findings demonstrate that subcostal ultrasonography combined with manometry is a feasible, valid and reliable method for assessing diaphragm contractile function across non-volitional, volitional and reflexive perturbations. With further refinement, this integrated method has translational potential for mechanistic research and clinical application.None
Air Pollution in 88 US Metropolitan Areas: Trends and Persistence
Full text linkData Availability Statement:
No new data were created or analyzed in this study.JEL Classification: C22; Q53; Q58.This paper analyses trends and persistence in air pollution levels in 88 US metropolitan areas using fractional integration methods. The results indicate that the differencing parameter d is higher than 0 in 38 of the series, which supports the hypothesis of long-memory behavior and implies that, although the effects of shocks are long-lived, they eventually die out. The highest degrees of persistence are found in the Fresno, Bakersfield, Bradenton and San Diego areas. On the whole, the gathered evidence indicates that regional differences in pollution levels are significant, with factors such as industrialisation history and extreme weather events playing a crucial role in their degree of persistence. This suggests that, in order to tackle pollution more effectively, federal environmental policies, such as the Clean Air Act, should be complemented by more targeted ones taking into account local characteristics.Prof. Luis A. Gil-Alana gratefully acknowledges financial support from the project from ‘Ministerio de Ciencia, Innovación y Universidades`Agencia Estatal de Investigación’ (AEI) Spain and ‘Fondo Europeo de Desarrollo Regional’ (FEDER), Grant D2023-149516NB-I00 funded by MCIN/AEI/ 10.13039/501100011033, and from an internal Project of the Universidad Francisco de Vitoria
<i>H</i><sub>∞</sub> Fuzzy Control for A Class of Cyber-Physical Systems Under Frequency-Duration-Constrained Replay Attacks
No full textIn this article, the H∞ fuzzy control problem is investigated for a class of nonlinear systems subject to replay attacks with frequency-duration constraints. Owing to the vulnerability of the open shared communication network, the information transmitted from the sensor to the controller may be exposed to replay attackers. A novel yet comprehensive replay attack model is constructed to characterize the repeated replay behavior of the adversary. On the basis of the constructed model, a fuzzy controller is designed to guarantee asymptotic stability and the desired H∞ performance. By employing Lyapunov stability theory and the orthogonal decomposition technique, sufficient conditions are derived to ensure the existence of the desired controller parameter. Finally, simulation results are presented to verify the effectiveness and correctness of the proposed fuzzy controller for T-S fuzzy systems under replay attacks.This work was supported in part by the National Natural Science Foundation of China under Grants
62273087 and 62503235; in part by the Startup Foundation for Introducing Talent of Nanjing
University of Information Science and Technology of China under Grant 1083142501018; in part by the
Fundamental Science (Natural Science) Research Project of Higher Education Institutions of Jiangsu
Province of China under Grant 25KJB413007; in part by the Royal Society of the UK; and in part by
the Alexander von Humboldt Foundation of Germany
Optimizing potential-based reward automata in partially observable reinforcement learning using genetic local search
No full textHighlights:
• Introduce a new RA learning mechanism with reward constraints for better strategies.
• Develop evolutionary algorithms to optimize RA and policies in various environments.
• Conduct experiments showing superior performance in six partially observable domains.
• Analyze exploration–exploitation balance and environmental randomness effects.
• Demonstrate the stability and efficiency of our genetic local search method.Data availability:
No data was used for the research described in the article.Partially observable reinforcement learning extends the reinforcement learning framework to environments in which agents have limited visibility of the state space, making it particularly relevant for applications in robotics and autonomous vehicle navigation. However, a primary challenge in partially observable reinforcement learning is defining effective reward functions that can guide the learning process despite partial observability. To address this challenge, this paper introduces a novel approach for constructing potential-based reward automata by employing genetic local search methods. Specifically, our method constructs these automata from compressed representations of exploration trajectories, which succinctly capture critical decision points and essential state transitions while eliminating redundant steps. By optimizing trajectory samples and shortening agent trajectories to their crucial transitions, our technique significantly reduces computational overhead. Formally, we define the learning objective as an optimization problem aimed at maximizing the log-likelihood of future observations while simultaneously minimizing the structural complexity of the learned reward automata. Furthermore, by incorporating value-based strategies to estimate potential values within the reward automata, our approach improves learning efficiency and facilitates the identification of optimal reward structures. We empirically evaluate our proposed method on seven partially observable grid-world benchmarks. Experimental results demonstrate that our method achieves superior performance relative to state-of-the-art reward automata-based techniques, exhibiting both accelerated learning speeds and higher accumulated rewards. Additionally, our genetic local search algorithm consistently outperforms comparative heuristic methods in terms of learning curves and reward accumulation.This work was supported by National Natural Science Foundation of China (No. 62202067), CNPC Innovation Found (No. 2024DQ02-0501), Royal Society (IEC_NSFC_233444), Postgraduate Research and Practice Innovation Project of Jiangsu Province (No. KYCX24_3228) and Youth Science and Technology Talent Promotion Project of Jiangsu Province (No. JSTJ-2025-137)
Analytical characterization of grid-forming converter response time under voltage sags
Full text linkThe response time of grid-forming (GFM) converters under grid disturbances is a critical metric that reflects how fast GFM capabilities can be delivered. In this letter, the analytical expression for the response time under voltage sags is derived using the geometric singular perturbation method. It is revealed that the response speed is naturally faster in weaker grids, and for a wide range of grid conditions, the response time under voltage sags has an inherent minimum limit of a quarter of the fundamental cycle, i.e., 5 ms at 50 Hz. Increasing the voltage-loop integral gain kᵢᵥ approaches this limit, but an excessively large kᵢᵥ induces small-signal instability. Hence, a transient reactive current augmentation strategy is proposed to approach and even break the limit while ensuring stability. Moreover, it is derived that current-limiting control reduces the response time. Therefore, the analytical response time constitutes an upper bound on the practical response time with current limiting. The findings suggest potential refinements to grid code requirements for GFM converter response time. The results are validated by hardware-in-the-loop testing.This work is supported in part by the National Key Research and Develop-ment Program of China under Grant 2024YFB2408600, in part by the National Natural Science Foundation of China under Grant 52322705, and in part by the special fund of Jiangsu Province for the Transformation of Scientific and Technological Achievements under Grant BA2023108
Data sources:
Access Repository Dashboard
Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇