30793 research outputs found
Sort by
State-of-the-Art and Challenges of Engineering ML- Enabled Software Systems in the Deep Learning Era
Get PDFEmerging from the software crisis of the 1960s, conventional software systems have vastly improved through Software Engineering (SE) practices. Simultaneously, Artiicial Intelligence (AI) endeavors to augment or replace human decision- making. In the contemporary landscape, Machine Learning (ML), a subset of AI, leverages extensive data from diverse sources, fostering the development of ML-enabled (intelligent) software systems. While ML is increasingly utilized in conventional software development, the integration of SE practices in developing ML-enabled systems, especially across typical Software Development Life Cycle (SDLC) phases and methodologies in the post-2010 Deep Learning (DL) era, remains underexplored. Our survey of existing literature unveils insights into current practices, emphasizing the interdisciplinary collaboration challenges of developing ML-enabled software, including data quality, ethics, explainability, continuous monitoring and adaptation, and security. The study underscores the imperative for ongoing research and development with focus on data- driven hypotheses, non-functional requirements, established design principles, ML-irst integration, automation, specialized testing, and use of agile methods
Press freedom and stock price crash risk
Get PDFData availability:
Data will be made available on request.This paper examines the impact of press freedom, an important institutional factor, on stock price crash risk. Using a large international sample of firms across 52 economies between 2002 and 2021, we find that firms in economies with higher degrees of press freedom are associated with lower levels of future stock price crash risk. Our analysis further shows that press freedom helps to deter the hoarding of bad news by increasing the intensity of reporting, extending the reporting period, and broadening local media coverage. Firms operating in economies with press freedom demonstrate stronger corporate governance and lower levels of firm-specific and long-term overvaluation, which are likely mechanisms through which press freedom mitigates crash risk. The negative impact of press freedom on crash risk is weakened by corruption but strengthened for firms facing higher short interest and less analyst coverage. Additional tests reveal that this negative relationship is driven by a collective influence from multiple dimensions of press freedom. Our results survive a battery of robustness checks. In sum, our findings suggest that press freedom enhances the stability of the global stock market by discouraging the concealment of negative information.Zhiyang Hui acknowledges financial support from Chongqing Social Science Project (China) (No. 2023BS021)
Estimating vertical land motion-adjusted sea level rise in a data-sparse and vulnerable coastal region
Get PDFData availability statement:
The data that support the findings of this study are available from the corresponding author upon reasonable request.Supplemental material is available online at: https://www.tandfonline.com/doi/full/10.1080/19475705.2025.2545375# .Sea level rise (SLR), driven by global warming, threatens coastal Bangladesh through inundation, land loss, and displacement. However, SLR estimates are often inconsistent or overestimated due to limited data and inadequate correction for vertical land motion (VLM). This study presents an integrated approach to accurately assess SLR by combining multi-station tide gauge (TG) records with satellite altimetry (SA) and interferometric synthetic aperture radar (InSAR) data across Bangladesh’s coastline. Relative SLR (RSLR) rates were derived from TGs, absolute SLR (ASLR) from SA, and InSAR-derived VLM trends were used to correct TG-based estimates. Results revealed strong seasonal variations, with sea levels peaking in April and lowest in September. Decadal trends indicated alternating phases of rise and fall. Annual SLR rates averaged 5.40 mm/year from TGs and 4.94 mm/year from SA, with notable spatial variations. VLM analysis showed subsidence at five TG sites and uplift at six. After VLM adjustments, all stations exhibited positive ASLR trends, averaging 4.58 mm/year. This study demonstrates that incorporating VLM and corrections of TG records significantly improves SLR estimation. The findings provided critical insights into the spatial and temporal dynamics of sea level change and provide a scientific basis for climate adaptation and infrastructure planning in Bangladesh’s vulnerable coastal zone.The authors would like to acknowledge the School of Earth and Planetary Sciences at Curtin University for providing research support grants. MSGA received support from the Leverhulme Trust through an Early Career Fellowship [grant reference ECF-2023-074]
Multi-scale Lipschitz Neural Fields Incorporating Frequency Decoupling for medical image registration
No full textData availability:
Data will be made available on request.Deformable image registration is a pivotal task in medical image analysis, essential for applications such as surgical navigation, lesion localization, and treatment planning. However, the accuracy of registration is frequently constrained by the ability to analyze complex textural features, which impedes the capture of fine-grained deformations and elevates surgical risks. To overcome these limitations, we propose the Multi-Scale Lipschitz Neural Fields Incorporating Frequency Decoupling (MLNFFD) framework, which reformulates the registration task as the correction of high-frequency residuals based on a low-frequency coarse registration foundation. Specifically, we employ a dual-branch architecture, comprising a band-limited U-Net low-frequency branch for coarse deformation estimation, and a Multi-Scale Lipschitz Neural Fields (MLNF) high-frequency branch dedicated to refining fine-grained deformation residuals. Through a complementary joint optimization strategy, the low-frequency branch provides an initial estimation that accelerates the optimization of the high-frequency branch, while the high-frequency branch refines the deformation field and enhances local details, fostering mutual reinforcement and improving registration accuracy. Additionally, the MLNF branch incorporates a Multi-Scale Implicit Dual-Domain Feature Representation (MIDFR), a deep similarity prior, and Lipschitz continuity constraints, enhancing the representation of complex textural features while ensuring the smoothness of the deformation field. Extensive experimental results on brain MRIs and cine cardiac MRIs datasets show that our proposed approach outperforms existing methods in terms of both registration accuracy and deformation field smoothness.This work was supported by the National Natural Science Foundation of China under Grant 62471077
Privacy-Preserving Bidirectional Data Transmission of Smart Grid via Semi-Quantum Computation: On Mutual Identity and Message Authentication
Get PDFThis paper is concerned with the privacy-preserving bidirectional electric power data transmission problem of the smart grid. A privacy-preserving bidirectional data transmission (BDT) protocol is developed over semi-quantum computation with aim to achieve bidirectional sensitive data flow between power suppliers and users. The minimal quantum cost is pursued under practical constraints while ensuring that power sensitive information is not leaked. To achieve the goal, a two-way data transmission protocol is first proposed that combines mutual identity authentication with message authentication for the benefits of enhanced security. Furthermore, for the preservation of privacy, a semi-quantum duplex communication approach is utilized, wherein the quantum state is randomly divided into two parts: teleportation and measurement qubits. The effectiveness of the privacy-preserving scheme against existing attack strategies is also rigorously analyzed. Lastly, simulation studies conducted on the IBM quantum cloud platform validate and underscore the superiority of the developed privacy-preserving BDT protocol.10.13039/501100004735-Natural Science Foundation of Hunan Province (Grant Number: 2024JJ5273);
10.13039/501100001809-National Natural Science Foundation of China (Grant Number: 62272483);
European Union’s Horizon 2020 Research and Innovation Programme (Grant Number: 820776);
Royal Society of the U.K.;
Alexander von Humboldt Foundation of Germany
A Multi-Objective Genetic Programming with Size Diversity for Symbolic Regression Problem
Get PDFGenetic programming has been positioned as a fit-for-purpose approach for symbolic regression. Researchers tend to select algorithms that produce a model with low complexity and high accuracy. Multi-objective genetic programming (MOGP) is a promising approach for finding appropriate models by considering tradeoffs between accuracy and complexity. The MOGP has gained significant attention for non-dominated sorting genetic algorithm II (NSGA-II). However, NSGA-II tends to excessively select individuals of lower complexity, making NSGA-II inefficient in real world applications. SD can be a strategy to promote the evolutionary process by adapting selection pressures for individuals of various size. It deals with the excessive tendency to select low complexity individuals in NSGA-II.We also introduce a practical industrial case of defect detection for dispensing machines. By modeling the dispensing volume of the fluid dispensing systems, defects in the dispensing machine can be detected under different external environmental factors.For the validation of SD, other MOGP algorithms are compared with the improved NSGA-II algorithm, NSGA-II with SD. By comparing multi-objective optimization methods tested on seven general datasets and an industrial case about defect prediction, the experimental results show that performance of the proposed approach is superior or same to other models in terms of accuracy. In terms of complexity, performance of the proposed approach is satisfactory.10.13039/501100001809-National Natural Science Foundation of China
Experimental Characterization of a Commercial Photovoltaic Thermal (PVT) Hybrid Panel Under Variable Hydrodynamic and Thermal Conditions
Get PDFData Availability Statement:
Data are contained within the article.Photovoltaic thermal (PVT) hybrid systems offer a promising approach to maximizing solar energy utilization by combining electricity generation with thermal energy recovery. This study presents an experimental evaluation of a commercially available PVT panel, focusing on its thermal performance under varying inlet temperatures and flow rates. The work addresses a gap in the literature regarding the real-world behavior of integrated systems, particularly in residential settings where space constraints and energy efficiency are crucial. Experimental tests were conducted at three mass flow rates and five inlet water temperatures, demonstrating that lower inlet temperatures and higher flow rates consistently improve thermal efficiency. The best-performing condition was achieved at 0.012 kg/s and 10 °C. These findings deepen our understanding of the panel’s thermal behavior and confirm its suitability for practical applications. The experimental platform developed in this study also enables standardized PVT testing under controlled conditions, supporting consistent evaluation across different settings and contributing to global optimization efforts for hybrid solar technologies.This research was funded by Brunel University London and Universidad UTE. The financial support included contributions from Universidad UTE (USD 2000) and the Ecuadorian Government (USD 5000) under a scholarship awarded through an open call
Drivers and barriers to rural and urban healthcare placement in Ghana: a Delphi study
Get PDFData availability statement:
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.Objective: This study explored the level of consensus on the drivers and barriers influencing doctors’ decisions to work in rural versus urban areas. The study provides insights into systemic issues affecting healthcare workforce distribution in Ghana. Access to medical care is particularly important given the changing demographics of Ghana, including the growth of the older and chronically ill population and the high proportion of older adults living in rural areas.
Methods: A three-round e-Delphi study was conducted among doctors and regional directors of the Ghana Health Service using a seven-point Likert scale. A median score of ≥6 and an interquartile range of ≤1 was used as cutoffs. In total, 47 experts participated in the study. Although 55 initially registered interest, only 47 took part in the first round. By the second and third rounds, 42 experts remained engaged in the study.
Results: Experts reached consensus on 40 descriptors (78%), of which 37 (93%) were considered important. Doctors reached consensus on 11 and 7 important drivers and barriers of rural incentive adoption, respectively, while reaching consensus on 8 important drivers of urban incentive factors. Regional directors reached consensus on 4 and 7 important drivers of rural and urban factors, respectively. Four categorical themes emerged from the analysis. These are financial, professional development and career advancement, work-life balance, and community lifestyle factors.
Conclusion: The contrast in drivers and barriers between rural and urban healthcare workers necessitates tailored policy approaches, resource allocation strategies, and workforce planning efforts to ensure equitable access and quality care across diverse settings and among different sub-populations, especially the growing number of aged and chronically ill.The author(s) declare that no financial support was received for the research and/or publication of this article
Characterization of Cu-Nb-Cu heterostructure fabricated by high-pressure torsion
Get PDFHigh-pressure torsion (HPT) processing disrupts the thermodynamic equilibrium in immiscible systems and often produces nonequilibrium microstructures with unique properties. This study investigates the microstructural evolution and mechanical behaviour of a Cu-Nb immiscible alloy subjected to HPT under 6 GPa compressive stress. The HPT processing was performed on stacked Cu-Nb-Cu layers by up to 200 turns and this produced mechanically alloyed, homogenized disks free of porosity or cavities. Microstructural characterization using X-ray diffraction and scanning electron microscopy, coupled with energy-dispersive X-ray spectroscopy, revealed a stepwise evolution, including the reduction of segregation layers, the formation of nonequilibrium Cu-17 at%Nb solid solution in the disc processed at 200 HPT turns and an increased Nb insertion into the Cu lattice. Additionally, grain refinement and residual strain increments were observed with increasing torsional turns. Thereafter, the mechanical properties were evaluated using hardness mapping and tensile testing. The material exhibited strain hardening behaviour and achieved an ultimate tensile strength (UTS) exceeding 1.25 GPa. Following post-deformation annealing, the UTS decreased to ∼700 MPa due to recrystallization and recovery. These results provide a preliminary understanding of microstructural transformations and their impact on the mechanical properties of immiscible systems subjected to extreme deformation
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! 👇