IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
Resilient Supply Chain Network Planning Method with Two-Stage Stochastic Programming: Extension to Multiple Product Supply Chains
No full textPart 1: Smart and Sustainable Supply Chain Management in the Society 5.0 EraInternational audienceIn supply chain management, risk management is crucial to ensure a stable product supply while considering economic efficiency. It is essential to design resilient supply chains that can maintain production capacity by preparing for and responding to predictable risks. Our group has proposed strategic planning methodologies for the selection of appropriate material suppliers and the optimization of inventory levels across the supply chain, including suppliers, manufacturers, and wholesalers, with considering potential risks. We have introduced a planning method for resilient supply chain networks using two-stage stochastic programming. This paper extends the proposed method to accommodate supply chains managing multiple products, aiming for more realistic conditions. The effectiveness of the extended method is assessed through computational experiments
Implementing a NFT Based Album Purchasing Web Application Using ReactJs by Integrating with Ethereum Blockchain
No full textPart 3: SDG 9 Industry, Innovation and InfrastructureInternational audienceBlockchain is one of the emerging technologies, which is trustless, decentralized and secure. In today’s world blockchain can be almost applicable anywhere. Using blockchain technology in the music industry, where many artists lose their music ownership and revenue to streaming platforms and production houses. Blockchain can be utilized to overcome this problem faced by artists, by using the concept of Non Fungible Token (NFTs). A dynamic web application built using ReactJs for the artist to mint their NFTs and store the metadata in off-chain using IPFS protocol and an ERC721 smart contract built using Solidity is deployed to Ethereum Blockchain and the IPFS CID along with the contract address is stored in MySql database using SpringBoot data Jpa and RESTful API for rendering data in webpage. Graph protocol is used for indexing and querying smart contract transactions and it is visualized using ChartJs
AI-Enhanced Sign Language Interpreter
No full textPart 2: SDG 4 Quality EducationInternational audienceCommunication is a basic human right, yet people with hearing impairments often face great challenges when trying to interact with the world. This project addresses the communication-demanding situations confronted with the aid of individuals with listening impairments by proposing a complete answer for hand signal gesture detection and translation into text and speech. Leveraging present-day technologies such as MediaPipe, OpenCV, and neural networks, the gadget captures actual hand gestures via a webcam. The neural community is designed and skilled for strong key point detection, as it should identify and classify the hand symptoms. The integration of OpenCV enables seamless interplay with the webcam, ensuring efficient gesture recognition. Upon successful detection, the corresponding gestures are translated into each text and speech, presenting a dual-mode communication interface. The utilization of neural networks complements the gadget's ability to evolve to a wide variety of gestures, selling inclusivity for customers using distinct sign languages. Upon successful detection, the system translates the identified gestures into both textual and spoken output, offering a dual mode communication interface. The utilization of neural networks enhances the system's adaptability to a diverse range of gestures, promoting inclusivity for users employing various sign languages. Notably, the proposed solution demonstrates an impressive accuracy of 97%, underscoring its efficacy in accurately interpreting and translating hand signals. The proposed solution now not simplest serves as a useful resource for the deaf and listening-to-impaired network but also establishes a basis for destiny advancements in assistive technology. This project contributes to the intersection of computer vision, system mastering, and accessibility, fostering innovation in the pursuit of an inclusive digital society
Assessing Degradation Levels of Palm Leaf Manuscripts with Random Forest Using Gabor Features
No full textPart 2: SDG 4 Quality EducationInternational audienceThis experiment focuses on employing machine learning techniques to create an automated system for classifying palm leaf manuscripts into three distinct categories based on their degradation levels: good, bad, and medium. The study collects a wide range of palm leaf samples at different degradation levels. Using advanced machine learning algorithms and an efficient feature extraction method, a model is created that accurately classifies palm leaves by level of degradation. The research incorporates Gabor feature extraction and statistical classifiers to train the extracted features. The classifiers used are k-Nearest Neighbor (k-NN), Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), Logistic Regression (LR), Random Forest (RF), Decision Tree (DT), Naive Bayes (NB), XG Boost (XGB), and Ada Boost (AB). After employing SMOTE, normalization, and feature selection, RF obtained the highest mean accuracy of 87.25% and a standard deviation of 0.08 in comparison with the other classifiers
Privacy Policies on Websites: A Case Study in the Financial Industry in South Africa
No full textPart 2: PrivacyInternational audienceFinancial companies handle clients’ personal data and outline the related data processing conditions in online privacy policies. Despite data privacy laws, legislative guidelines for these policies are lacking. This research aimed to address the current state of online privacy policies, firstly by proposing holistic criteria for what must be included in a website privacy policy. Thereafter, a case study methodology was applied to review online privacy policies in the financial industry in South Africa, applying the proposed criteria. The key findings of this research indicate that financial companies do not fully address the criteria relating to the data subject's choice and consent, the integrity and security of the data, enforcement and redress, and any further information transfer that may occur in terms of their online privacy policies. The proposed criteria offer value to financial companies by providing a measure by which to assess their online privacy policy content
A Holistic Approach to Developing Intervention Strategies Against Digital Piracy
No full textPart 2: PrivacyInternational audienceAddressing the global challenge of digital piracy, this study concludes a research series that explores the multifaceted drivers behind copyright infringement activities. Integrating findings from a PRISMA-guided systematic literature review and the application of behavioural psychology through the Theoretical Domains Framework (TDF), this work identifies key factors of digital piracy, including accessibility, awareness, education and social and cultural influences, alongside a consideration of previous behaviour. Crucially, the research leverages expert reviews analysed through ATLAS.ti, enhancing the development of the Digital Piracy Conceptual Framework (DPCF). The study’s findings, derived from interviews with one (1) participant from each of the six (6) sectors, provide sector-specific insights that, while informative, should not be interpreted as broadly generalisable across industries. This detailed approach has refined the DPCF, offering a comprehensive blueprint for devising effective digital piracy intervention strategies, marking a significant step towards mitigating this pervasive issue and protecting intellectual property rights globally
Discerning Challenges of Security Information and Event Management (SIEM) Systems in Large Organizations
No full textPart 4: Usable SecurityInternational audienceSecurity Information and Event Management (SIEM) systems are essential for security experts in various daily tasks such as monitoring, anomaly detection, forensics, identifying indicators of compromises, threat hunting, and incident handling. Although many different SIEM systems are being used in large organizations, there needs to be more understanding of the existing challenges of SIEM systems from a human-centric cybersecurity perspective. The present study explores those challenges following a qualitative research approach utilizing the Delphi technique. Two rounds of interviews were conducted with twelve security experts in multiple large organizations. The experts expressed the challenges in the first round, exploring various components of user, usage, and usability of SIEM systems. Then, the challenges were divided into thirteen main categories based on the consensus level. In the second round, the experts validated and ranked the categories. Results show that the most significant challenges are related to usage, followed by usability and user components
Research Agenda for Speaker Authentication
No full textPart 3: Technical Attacks and DefensesInternational audienceIn this study, we thoroughly examined every component of speaker authentication, analyzing the input, process, and output phases to identify flaws and new threats. Our investigation is organized around specific research topics that aim to effectively address and minimize the identified dangers. By methodically exploring each component of the speaker authentication process, we not only identify possible issues but also recommend proactive methods to protect these systems from unauthorized access. Our research questions act as significant probes, allowing for a deeper knowledge of the underlying difficulties and leading to the creation of tailored authentication solutions. This study goes beyond theoretical analysis and provides practical insights and strategic recommendations for improving the security and reliability of speaker authentication systems in a variety of sectors, including cybersecurity and forensic analysis. We highlight the interrelated nature of the input, process, and output stages, emphasizing the importance of remaining vigilant in the face of emerging security risks. Our goal is to provide the necessary knowledge and tools to effectively handle the complexities of speaker authentication in the changing digital world. This work establishes a solid foundation for the development of safe and durable speaker authentication methods
A Profile-Based Cyber Security Readiness Assessment Framework at Country Level
No full textPart 1: Management and RiskInternational audienceContinuous improvements to national cybersecurity policies are necessary due to the rapidly evolving cyber threat landscape, growing reliance on information and communication technologies (ICTs), and the prevalence of digital dangers. The process of evaluating cybersecurity maturity is becoming more and more crucial in the dynamic digital environment. Governments can do a thorough assessment of a nation's cybersecurity capabilities by using a cybersecurity maturity model. This allows them to pinpoint areas of weakness and offer specific recommendations for strengthening cybersecurity capabilities. These assessments serve as a standard for illustrating a country's readiness for cyberattacks (where cyber readiness refers to the organisation’s ability to identify, prevent and respond to cyber threats). However, the results of these maturity assessment models are not sufficient for creating national cyber security policies since they are overly generalised and deficient in the evaluation of present capabilities. The focus of this study is to propose a model that takes into consideration multiple factors having greater relevance to national cyber security strategies. This model introduces a profiling approach that carefully evaluates the country's readiness based on the current state of the cyber security initiatives taken at the national level. The key areas considered in the evaluation include threat landscape, overall cyber security posture, initiatives, legal frameworks, infrastructure support, collaborations, capacity, and workforce-building initiatives. To find the utilisation of the novel profiling approach, we have applied it to four selected countries that are recognised as potential targets for threats due to their growing internet population and connectivity. Our research outcome reveals that the profiling techniques reflect the current state of readiness at the national and organisational level to a greater extent and are more optimised for cyber maturity assessment
Predicting Hospital Length of Stay Using Light Gradient Boosting Machine Regression
No full textPart 3: Applications of MLInternational audienceLength of stay (LoS) in a hospital is an important metric in the healthcare management system, with profound implications for resource allocation, patient outcomes, and cost reduction. This paper reviews the literature on approaches for predicting hospital length of stay with a R2 score of 96.14 per cent using Light Gradient Boosting Machine regressor. We critically assess the merits and limitations of various methods and propose a unified framework for the generalized prediction of length of stay. Our framework includes investigating the types of routinely collected data and recommendations for robust knowledge modelling