30793 research outputs found
Sort by
Large AI Model-Based Semantic Communications
Get PDFThe article version on this institutional repository is available at arXiv:2307.03492v2 [cs.AI], https://arxiv.org/abs/2307.03492 (Sat, 3 Aug 2024 13:59:24 UTC (14,879 KB)).The source code of this article is available at: https://github.com/jiangfeibo/LAMSC.git .Semantic communication (SC) is an emerging intelligent paradigm, offering solutions for various future applications like metaverse, mixed reality, and the Internet of everything. However, in current SC systems, the construction of the knowledge base (KB) faces several issues, including limited knowledge representation, frequent knowledge updates, and insecure knowledge sharing. Fortunately, the development of the large AI model (LAM) provides new solutions to overcome the above issues. Here, we propose a LAM-based SC framework (LAM-SC) specifically designed for image data, where we first apply the segment anything model (SAM)-based KB (SKB) that can split the original image into different semantic segments by universal semantic knowledge. Then, we present an attention-based semantic integration (ASI) to weigh the semantic segments generated by SKB without human participation and integrate them as the semantic-aware image. Additionally, we propose an adaptive semantic compression (ASC) encoding to remove redundant information in semantic features, thereby reducing communication overhead. Finally, through simulations, we demonstrate the effectiveness of the LAM-SC framework and the possibility of applying the LAM-based KB in future SC paradigms.10.13039/501100001809-National Natural Science Foundation of China (Grant Number: 41904127,62132004). This work was supported in part by the National Natural Science Foundation of China under Grant 41904127 and 62132004, in part by the Hunan Provincial Natural Science Foundation of China under Grant 2024JJ5270, in part by the Open Project of Xiangjiang Laboratory under Grant 22XJ03011, and in part by the Scientific Research Fund of Hunan Provincial Education Department under Grant 22B0663
Non-fungible tokens and liability of online marketplaces: a European perspective
Get PDFThe rise of NFTs seems to open new horizons for the exploitation of works in the digital world. This, of course, does not come without challenges. A number of IP infringement disputes are now coming to the forefront; minting NFTs without the consent of the creator of the original work and malicious use of a trade mark as an NFT without the permission of the brand owners are but a few examples. In light of the absence of any judicial guidance, this chapter aims to reflect on how the use of NFTs can trigger copyright or trade mark infringements and examine to what extent NFTs marketplaces can be liable for IP infringements that take place within their platforms, taking into consideration a line of case law from the CJEU and EU legal instruments
Identifying and handling data bias within primary healthcaredata using synthetic data generators
Get PDFData availability:
The anonymised electronic healthcare record data used in this research is not publicly available but can be requested from CPRD subject to a data licence and research data governance (RDG) approval. The generated synthetic data set discussed in this paper can also be requested from CPRD subject to a data sharing agreement (DSA). Data access licence fees apply (https://cprd.com/data).
Code availability:
All our R code is available via GitHub (https://github.com/barbaraDraghi/BayesBoost). The R package bnlearn (v4.8.1) is used for all Bayesian network inference.Appendix A. Additional results are available online at: https://www.sciencedirect.com/science/article/pii/S2405844024001956#se0130 .Copyright © 2024 The Authors. Advanced synthetic data generators can simulate data samples that closely resemble sensitive personal datasets while significantly reducing the risk of individual identification. The use of these advanced generators holds enormous potential in the medical field, as it allows for the simulation and sharing of sensitive patient data. This enables the development and rigorous validation of novel AI technologies for accurate diagnosis and efficient disease management. Despite the availability of massive ground truth datasets (such as UK-NHS databases that contain millions of patient records), the risk of biases being carried over to data generators still exists. These biases may arise from the under-representation of specific patient cohorts due to cultural sensitivities within certain communities or standardised data collection procedures. Machine learning models can exhibit bias in various forms, including the under-representation of certain groups in the data. This can lead to missing data and inaccurate correlations and distributions, which may also be reflected in synthetic data. Our paper aims to improve synthetic data generators by introducing probabilistic approaches to first detect difficult-to-predict data samples in ground truth data and then boost them when applying the generator. In addition, we explore strategies to generate synthetic data that can reduce bias and, at the same time, improve the performance of predictive models.NHSX grant: BEIS Innovate Regulatory Pioneer Fund, project: "Using High-fidelity Synthetic Data as synthetic control arms and to boost sample sizes in clinical trials"
A comparative study on transport and interfacial physics of H2/CO2/CH4 interacting with H2O and/or silica by molecular dynamics simulation
Get PDFData Availability:
The data that support the findings of this study are openly available in figshare at https://figshare.com/articles/dataset/cc_gas-liquid-solid/24438541, Ref. 87, under a CCBY license.Copyright © 2024 Author(s). Underground H2 storage (UHS), i.e., injecting H2 into subsurface geological formation and its withdrawal when needed, is identified as a promising solution for large-scale and long-term storage of H2. In this study, molecular dynamics (MD) simulation was performed at a typical temperature 320 K with pressure up to 60 MPa to predict H2 transport properties and H2–H2O–rock interfacial properties, which are compared with those of CO2 and CH4. The MD results show that the CH4 profiles of property variations with pressure lie between those of H2 and CO2 and more comparable to CO2. The interaction of H2 with H2O/silica is much weaker than that of CH4 and CO2. It is found that the effect of H2 pressure on altering the water contact angle and interfacial tension is negligible under all conditions. Unlike the multi-adsorption layers of the confined CO2 and CH4, there is only one adsorption layer of H2 confined by silica nano-slit. The planar diffusion of H2 in the confined system is slower than that in the bulk system at pressures lower than 20 MPa. The data and findings of this study will be useful for modeling the multiphase flow dynamics of UHS on reservoir scale, optimizing UHS operation, and assessing the performance of a cushion gas, e.g., CO2 or CH4.This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) under Grant No. EP/T033940/1. The authors are grateful to the high-performance computing (HPC) resources of ARCHER2 supported by the EPSRC Access to High Performance Computing under Project No. e774 and UK Materials and Molecular Modelling Hub for computational resources, which is funded by EPSRC (Grant Nos. EP/T022213/1, EP/W032260/1, and EP/P020194/1)
A comprehensive review of renewables and electric vehicles hosting capacity in active distribution networks
Get PDF© Copyright 2023 The Author(s). The excessive integration of renewable distributed generation (RDG) and electric vehicles (EVs) could be considered the two most problematic elements representing the greatest threat to the distribution network (DN) technical operation. In order to avoid going beyond technical limitations, the term hosting capacity (HC) was proposed to define the highest permitted amount of distributed generation (DG) or EVs that can be integrated safely into the DN. The connection of RDGs was first brought to the attention of researchers and DN operators since it accounts for the most notable portion of these technical issues. Hence, the phrase ‘DG-HC’ was initially proposed and evolved significantly over the last few years. Currently, EV integration in most DNs worldwide is still low, but given the worldwide support for clean transportation options, expectations are raised for a significant increase. As a result, it is anticipated that over the next years, the effect of EV integration on the DN will be highly noticeable, requiring greater attention from researchers and DN operators to define the accepted limits of EV penetration levels, ‘EV-HC,’ which is expected to pass along the same line of DG-HC. This article provides an in-depth review of both DG-HC and EV-HC. It first analyses how the DG-HC research has grown over the years and then studies the published EV-HC papers, illustrating to what extent there is a similarity between them and, finally, employs these analyses to expect future development in the EV-HC research area. This article includes the different uses of the term HC, the most common performance indices of DG-HC, the various methods for assessing DG-HC, the different techniques for DG-HC enhancement, the effects of integrating EVs on the DG-HC, and finally, calculating and enhancing methods for EV-HC
Developing a game design framework to embed student-centred learning
Get PDFData Access Statement: All the data generated in this study is presented within the manuscript and the authors
encourage others to use and cite this game design framework.Purpose:
This paper presents the student-centred experience (SCE) game design framework, which aims to guide the design of holistic student-centred digital game-based learning (SCDGBL) experiences, which fully integrate all seven tenets of student-centred learning (SCL). The paper also rationalises the need for the framework and presents the steps taken in its development.
Design/methodology/approach:
Initially, the background areas of SCDGBL and digital game-based learning (DGBL) are examined, and the need for a framework in digital educational game design that has a focus on SCL is then established. The rigorous and systematic design thinking process through which the framework was developed is then stepped through. The completed framework is then presented, and each section is detailed to explain its utilisation within the process of digital game design.
Findings:
The paper presents the completed student-centred experience (SCE) framework alongside a worked example of how it can be deployed in practice. Also included is guidance on the roles of the game designer and education practitioner at all stages of design, development and deployment and how they may contribute their experience during the game design process to create high-quality tools for learning.
Research limitations/implications:
While the SCE framework presented is complete, it is presented as a first version and will benefit from wider deployment and testing.
Originality/value:
This paper presents a new game design framework integrating existing knowledge on SCL and DGBL, which guides practitioners in the design of experiences that fully deliver the techniques of both areas.This work was funded by the Engineering and Physical Sciences Research Council as part of a doctoral studentship
Social Value at the Heart of the Community in the Context of Urban Regeneration: from meanwhile resilience to community sustainability
Get PDFThe Grahame Park and Colindale Community Research Project was funded by Colindale Communities Trust (see: https://www.brunel.ac.uk/research/projects/social-value-in-urban-regeneration).This report summarises the results of a programme of research which took place at the Colindale Communities Trust (CCT) between January and July 2023. The CCT is a community-based charity situated on the Grahame Park Estate which is undergoing regenera on but meanwhile suffering from physical and social neglect. They currently manage the Old Library community hub on Grahame Park. Although catering for the wider Colindale community, The CCT is at the heart of the Grahame Park Estate and Grahame Park remains one of the 5 areas with the highest level of depriva on in Barnet.Colindale Communities Trus
A novel cohesive interlayer model considering friction
Get PDFData Availability: The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.To understand the influence of friction on the shear-slip behavior of heterogeneous brittle composites, a novel cohesive interlayer model that can effectively capture the friction effect was proposed based on the classical Park-Paulino-Roesler model. Meanwhile, the unified potential energy function governing the interface tangential and normal behaviors was introduced to realize the mechanical interaction between Mode I fracture and Mode II fracture, and a smooth friction growth function was added in the elastic deformation stage for calculating the accurate contact pressure and friction force. Furthermore, the capability of the proposed model in addressing unloading and reloading was improved, and the fracture energy can vary accordingly during cyclic loading. To verify the effectiveness of the proposed model, it was examined by modelling the shear behavior of a masonry wallette. The results show that the relative error of the proposed model is 14.92% which is much lower than those of the other three pre-existing models when calculating the displacement corresponding to peak shear stress. Meanwhile, in terms of peak shear stress and initial displacement at residual stage, the relative errors of the proposed model are only 1.82% and 5.04%, respectively, indicating the high accuracy. Besides, the tangent stiffness determined by the second-order integration of the potential energy function is also continuous and smooth, which ensures the effective convergence of the proposed cohesive model.National Natural Science Foundation of China (Grant No. 41941018)
The distributional effects of climate change. An empirical analysis
No full textSupplementary data are available online at: https://www.sciencedirect.com/science/article/pii/S0014292124001570?via%3Dihub#appSB .This paper benefited from comments by two anonymous referees, the editor Evi Pappa, and participants at the ‘Climate Change and the Global Economy’ workshop at the Lancaster Business School and at the following conferences: SETA2023, CRETE and IAAE 2022. A previous version of this paper was circulated with the title: ‘Climate change and income inequality. An empirical analysis’ (Queen Mary University of London. School of Economics and Finance Working Paper No. 966, available at: https://www.econstor.eu/handle/10419/284316).The role of climate change on output has been studied extensively in the empirical literature. However, its distributional implications have received little attention. This paper attempts to fill this gap by investigating if climate shocks affect income inequality. Using a Vector Autoregression for a large cross-country panel, we identify the climate shock in the frequency domain as the shock that explains the bulk of the variance of climate variables in the long-run. An adverse climate shock is associated with an increase in measures of income inequality, affecting mostly low income households. The impact of the shock is larger in magnitude for low income, hot countries with a significant agricultural sector and low degree of adaptation to climate change
Enhancing healthcare facility resilience: utilizing machine learning model for airborne disease infection prediction
Get PDFDuring this pandemic, advanced epidemiological models have been widely used to determine intervention strategies for controlling the spread of the disease in public and healthcare settings. These models played a crucial role by providing predictive insights into disease transmission dynamics, informing resource allocation and guiding policy decisions. However, the accuracy of these predictions depends on the substantial amount of input data, which was not readily available at the onset of the pandemic. Another concern with the existing models is their inability to adequately account for the complex indoor built environments, which has been shown to significantly impact infection risk. To tackle these issues, this paper discusses the potential of developing a joint modelling technique that integrates machine learning models, building information models and agent-based models to assess the risk of nosocomial airborne infections. With limited available data, machine learning models can determine infection risk with high confidence
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! 👇