1453 research outputs found

    CERN–UNESCO–NFR School on Open Science

    No full text

    DRC CONFLICT: SECOND MESSAGE TO WORLD LEADERS

    Full text link
    Second message to World Leaders following the 'Petition for the dissolution of the United Nations and the prosecution of certain personalities and their associates for the crime of genocide in the Democratic Republic of Congo' which was sent to them in October 2024

    DEEP NEURAL NETWORKS FOR BIOMETRIC AUTHENTICATION IN INFORMATION SECURITY

    Full text link
    In the rapidly evolving landscape of information security, biometric authentication has emerged as a critical mechanism for enhancing identity verification processes. Deep Neural Networks (DNNs), with their exceptional capability to learn complex patterns and representations from large datasets, have significantly advanced the accuracy and robustness of biometric systems. This paper explores the application of DNNs in biometric authentication, focusing on modalities such as fingerprint recognition, facial recognition, iris scanning, and voice authentication. We discuss the architectural innovations in deep learning, including Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), that contribute to improved feature extraction, noise reduction, and resistance to spoofing attacks. Furthermore, the paper addresses the challenges associated with deploying DNN-based biometric systems, such as data privacy concerns, model interpretability, and susceptibility to adversarial attacks. By analyzing recent advancements and experimental results, we demonstrate how DNNs enhance the security, scalability, and efficiency of biometric authentication in modern information security infrastructures

    NEURAL NETWORK ARCHITECTURES FOR SECURE FEDERATED LEARNING

    No full text

    Interpreting access to the Dinaledi Subsystem by Homo naledi: redefining the “Chute” as a labyrinth

    No full text
    The first detailed study and mapping of the access and egress routes into and out of the Dinaledi Chamber, Rising Star Cave, South Africa and its implications for understanding the use of the space by Homo naledi.A large assemblage of fossils attributed to the hominin species Homo naledi was originally discovered in the remote and difficult-to-access Dinaledi Subsystem of the Rising Star cave system, South Africa. Entry into this section by scientists and explorers has always been through a narrow and complex fissure network that includes a substantial vertical descent. This area was named the “Chute” and was described and illustrated in simplified form as a ~12 m vertical passage that gives direct access from the entry point into the chamber itself. However, due to the very constrained nature of the fissure network, previous descriptions of this space have not been based on detailed survey. In this paper, we provide the first detailed survey data and accurate spatial descriptions of the area between the Dragon’s Back and the Dinaledi Subsystem, showing the full complexity of the horizontal and vertical fractures and fissures that comprise the network connecting these spaces. To reduce misconceptions, we have renamed this space the “Chute Labyrinth” to highlight its true form. While some changes have occurred in the last 335 ka, these did not change the access from the rest of the cave into the Dinaledi Subsystem. The new spatial data and re-description of the space contextualize how Homo naledi individuals physically entered the Dinaledi Chamber via the Chute Labyrinth into the chambers below to dispose of their dead.The National Geographic Society, John Templeton Foundation, Lyda Hill Foundatio

    HYBRID NEURAL NETWORK MODELS FOR PRIVACY-PRESERVING DATA PROCESSING

    Full text link
    In the era of big data, the need to balance efficient data processing with robust privacy safeguards has become increasingly critical. This paper explores Hybrid Neural Network Models as a promising solution for privacy-preserving data processing. By integrating the strengths of traditional neural architectures (such as Convolutional Neural Networks and Recurrent Neural Networks) with privacy-enhancing technologies like differential privacy, federated learning, and homomorphic encryption, hybrid models can achieve high performance without compromising sensitive information. We discuss architectural designs that optimize both data utility and privacy protection, focusing on adaptive learning mechanisms that minimize data exposure during training and inference. Experimental evaluations across diverse datasets demonstrate that these models maintain competitive accuracy while significantly reducing privacy risks. This research highlights the potential of hybrid neural networks to enable secure, scalable, and efficient data processing in privacy-sensitive domains such as healthcare, finance, and IoT

    The Role of COAR in Advancing Open Access in Africa

    Full text link
    This webinar session featured Kathleen Shearer, the Executive Director of the Confederation of Open Access Repositories (COAR), as our guest speaker. She spoke on the topic: The Role of COAR in Advancing Open Access in Africa In this webinar, attendees discovered: - How COAR supports open access to research in Africa. - Ways African researchers and institutions can make their work more visible worldwide. - Why collaboration and global standards are important for African repositories. - Exciting programs and opportunities COAR offers to support African research. This webinar series is co-organized by: UbuntuNet Alliance: https://ubuntunet.net/ and Access 2 Perspectives: https://access2perspectives.org/ and supported by the African Open Science Platform (AOSP) https://aosp.org.za/The Role of COAR in Advancing Open Access in Africa - AfricArXiv Open Science Webinar Series 202

    Plant-Derived Compounds with Therapeutic Potential for the Treatment of Human Coronavirus Diseases

    Full text link
    Since the advent of modern civilization, few diseases have caused more worldwide socioeconomic disruptions than COVID-19 caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2). In some countries, very high proportion of the population have contracted the disease even though it is fatal in only a small percentage of those carrying the virus. In addition to physical measures (social distancing, handwashing, sterilization, etc.), two mainpharmaceutical interventions have been adopted: the development of vaccine/mass vaccination and therapeutic intervention. The recent discovery of SARS-CoV-2 transmission among vaccinated individuals and changes in the genome of the virus arising from mutation leading to new variants (delta, omicron, IHU etc.) suggest the importance of therapeutic intervention targeting different aspects of the molecular mechanisms involved in its virulence. Therapeutic agents targeting essential elements required for viral propagation in the host may also nd application in the management of other viral diseases such as Ebola, Zika, and HIV/AIDS. Targets for drug design include the 16 non-structural proteins, RNA-dependent DNA polymerase, esterase, membrane proteins, spike and envelope proteins, protease and nucleocapsid proteins, and helicase, all present on the virus; host proteases and receptors. Both medicinal plant-derived and synthetic compounds including monoclonal antibodies are now suggested as candidate drugs for COVID-19 and are being developed as suitable therapeutic agents. In this review, some useful information on promising plant-derived therapeutic agents are provided which may be of value in the development of drugs for COVID-19 and other viral diseases. 6/26/24, 6:14 PM Plant-Derived Compounds with Therapeutic Potential for the Treatment of Human Coronavirus Diseases | Proceedings of the National Academy of Sciences, India Section B: Biologic… https://link.springer.com/article/10.1007/s40011-024-01658-5 2/1

    Institutional Repositories serve the interests of research funders

    Full text link
    Presentation held at the Africa PID Alliance (africapidalliance.org) Webinar: How effective FAIR data management aids in the visibility of African funding output "Effective FAIR data management significantly aids in the visibility of African funding outputs by making research more discoverable, accessible, interoperable, and reusable. This not only boosts the profile of African research on the global stage but also fosters collaborations, enhances credibility, informs policy, and attracts further funding. For more on how we have been doing this please join our partners Crossref and Research Organization Registry(ROR) as we take you through what we have been doing and listen to some of the African funders case studies from National Research Fund Kenya and National de la Recherche et de l’Innovation pour le Développement - FONRID (Burkina Faso) in addition with the contributions from Association of African Universities and African Open Science Platform who will share the research management and policy-making implications on making African funders output accessible." Presentation held at the Africa PID Alliance (africapidalliance.org) Webinar: How effective FAIR data management aids in the visibility of African funding output "Effective FAIR data management significantly aids in the visibility of African funding outputs by making research more discoverable, accessible, interoperable, and reusable. This not only boosts the profile of African research on the global stage but also fosters collaborations, enhances credibility, informs policy, and attracts further funding. For more on how we have been doing this please join our partners Crossref and Research Organization Registry(ROR) as we take you through what we have been doing and listen to some of the African funders case studies from National Research Fund Kenya and National de la Recherche et de l’Innovation pour le Développement - FONRID (Burkina Faso) in addition with the contributions from Association of African Universities and African Open Science Platform who will share the research management and policy-making implications on making African funders output accessible." Watch the recording at https://doi.org/10.21428/359fb1d2.0361a9d

    Enhancing the Policy Impact of African Research: Strategies for Visibility and Influence through Overton

    Full text link
    This webinar session featured Euan Adie, the CEO of Overton. He spoke on the topic: Enhancing the Policy Impact of African Research: Strategies for Visibility and Influence through Overton African researchers and other attendees learned how to make their research more visible to policymakers and track its impact using Overton. They also got tips on engaging with policymakers and saw examples of how African research has influenced policy decisions. This webinar series is co-organized by: UbuntuNet Alliance: https://ubuntunet.net/ and Access 2 Perspectives: https://access2perspectives.org/ as part of the ORCID Global Participation Program: https://info.orcid.org/global-participation-program/Enhancing the Policy Impact of African Research: Strategies for Visibility and Influence through Overton - AfricArXiv Open Science Webinar Series 202

    540

    full texts

    1,453

    metadata records
    Updated in last 30 days.
    AfricArXiv
    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! 👇