IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
Information Security and Privacy Challenges Related to Personal eHealth Services - A Literature Review
No full textPart 1: Management and RiskInternational audienceThe study aimed to identify information security and privacy challenges concerning Personal eHealth Services (PeHS) via a systematic literature review. The result shows that there are several challenges to consider. In total, 8 themes of challenges were identified. Some examples of challenges are access control, patient trust, collaboration between multiple parties and the need for more knowledge. Further, to deal with the challenges, there is a need to improve governance and management of information security and privacy at the regional and national levels to include new services such as PeHS. Regardless of whether the patient information goes via the intra/inter-organizational e-health services or the Patient eHealth Services, the information is part of the patient's total information flow and must be included as a prominent part of healthcare's information security and privacy work to gain patient-centred and transparent care
Multi-factor Authentication Adoption: A Comparison Between Digital Natives and Digital Immigrants in Sweden
No full textPart 4: Usable SecurityInternational audienceMulti-Factor Authentication (MFA) is commonly suggested as a good mechanism to overcome inherent security problems with the use of passwords. However, research suggests that MFA has so far failed to attract enough interest from users. Additionally, older users seem to be even more reluctant to use MFA. In Sweden, users are more or less required to use MFA to use services such as online banking, book doctors appointments online, and complete tax reports online. As such, Sweden is an interesting case for studying MFA adoption. This paper reports on mixed-methods research investigating how Swedish users in different age groups compare with respect to the adoption of MFA. The results suggest that users of different age are willing to adopt MFA when it is required for services they want or need to use. However, younger users appear to be more prone to voluntarily adopt MFA
Campus Placement and Salary Prediction: Leveraging Machine Learning for EnhancedEmployability
No full textPart 2: Data AnalyticsInternational audienceIn an era of fierce talent competition, optimizing campus placement and predicting graduate salaries is vital. This paper explores ‘Campus Placement and Salary Prediction’ using SVM, Random Forest, Logistic Regression, KNN, and Gradient Boosting. We analyze a comprehensive dataset with student info, academics, skills, internships, and placement outcomes to identify success factors. We employ a majority voting rule among these models and have created a user-friendly website for practical use in academic institutions. Our multifaceted research supports institutions in enhancing student employability and aligning academic goals with the evolving job market. Logistic Regression, one of the models employed has campus placement prediction, has an accuracy of 84%, whereas Gradient Boosting, one of the models used for salary estimation, has an accuracy of 78%
Supply Chain Management Using Optimization and Machine Learning Techniques
No full textPart 2: Data AnalyticsInternational audienceThis study focuses on developing a predictive model for late deliveries in the supply chain, using the Data Co Supply Chain dataset. It involves data cleaning, visualization, and training various machine learning algorithms, selecting the most effective model based on accuracy and recall values. The research explores challenges like transportation delays, production issues, natural disasters, and supplier disruptions. Businesses are adopting proactive supply chain and risk management strategies to enhance resilience. The paper discusses how machine learning techniques can optimize supply chain management, offering insights into mitigating risks associated with late deliveries. By identifying challenges in advance, businesses can improve operational efficiency, minimize financial setbacks, and maintain customer trust. In a dynamic business environment, this research provides valuable solutions for supply chain professionals, integrating big data and machine learning to forecast and address late deliveries. It not only aids in predicting delays but also contributes to proactive decision-making, improving overall supply chain performance amidst uncertainties
Real-Time Soil Moisture Sensing Using Arduino for Automated Plant Irrigation System
No full textPart 2: Data AnalyticsInternational audienceIrrigation constitutes a fundamental element of agriculture and food production. Outdated methods in both developing and underdeveloped nations result in significant water wastage during this crucial process. This project proposes an inventive method for automating the control of water release valves in micro irrigation networks with prediction of required humidity by weather condition using artificial intelligence (AI). This system seeks to close the gap between water supply and crop demand while also optimizing resource use and reducing environmental impact by using AI-powered soil moisture monitoring and decision-making processes. This project address the key factors like varying demand of water for crops based upon the growth stage, weather conditions and plant type. This reducing the usage by adopting the method of micro irrigation which reduces the usage of water by 70–80% than conventional methods of irrigation. This project also provides a user-friendly interface to efficiently control the watering of plants when needed by the user
Campus Drive Portal on Career Advancement for College Students
No full textPart 2: Data AnalyticsInternational audienceThe Campus Drive Portal is a user-friendly website designed to assist both students and Placement Admins in managing campus recruitment events. This application provides the facility to view upcoming campus drives. The two users of our application are Student and Placement Admin. From Placement Perspective, the portal simplifies the management of student information for upcoming drives. This includes creating, updating, viewing, and deleting student records, with the ability to filter based on criteria set by companies. Admin can update the details like Company’s name, drive date and eligibility criteria for the upcoming placement drives. The portal also incorporates a prediction tool to assess a student’s chances of securing placement and further the trend analysis is used to help students by providing better training programs. The placement administrator has the facility to collect the offer letters from the placed candidates, which will be helpful in storing the details and analyzing the trends. From a Students’ Perspective, Students can receive information regarding the upcoming drives. After interviews, students could contribute valuable feedback about their experiences with companies. This feedback not only benefits the individual student but also aids the placement department in making informed decisions about future collaborations with companies. The system further includes a prediction tool that allows students to evaluate their interview preparedness and offers guidance on improving their skills. Students can also gain valuable insights from their seniors about different interview rounds, promoting a collaborative learning environment
Machine Learning Powered Nutritional Guidance System Using Personalized Nutritional and Digital Health Record
No full textPart 3: Applications of MLInternational audienceThe combination of digital health records and nutritional data has opened up a new world of substantiated salutary recommendations in the data-driven healthcare period of the moment. The objective of this design is to develop a state-of-the-art system that provides substantiated salutary suggestions to people according to their specific health biographies. This revolutionary system will enable druggies to make informed food choices that are in line with their individual health objects, salutary preferences, and constraints by integrating rich nutritive data with full electronic health records along with nearby location to get it. By exercising sophisticated recommendation algorithms and a flexible feedback medium, the system will constantly acclimate itself in trouble to enhance stoner health results. Icing data sequestration and security is of utmost significance, and the design will misbehave with all applicable regulations and norms
Automation Xtreme - A Web Automation AI Tool
No full textPart 1: Applications of AI/ML in KDM, Cloud Computing & SecurityInternational audienceIn the swiftly changing realm of web application development, the need for effective and dependable testing solutions is paramount. This document presents a groundbreaking Automated Web Application Testing Platform crafted to simplify the testing procedure and amplify overall testing efficacy. Leveraging powerful technologies such as Selenium for web automation, Pytest for testing, and Twilio for user support, the platform offers a comprehensive solution for web application testing. The platform’s intuitive user interface streamlines the definition and execution of test cases, ensuring accessibility for users across diverse technical proficiency levels. A unique feature of dynamic code generation based on user-defined test cases allows for flexibility and adaptability in testing scenarios. The modular architecture ensures maintainability, scalability, and easy integration with other tools and frameworks. Extensive testing methodologies, including black box and white box testing, validate the platform’s robustness and reliability. The integration of Twilio enables real-time communication through SMS notifications, enhancing user engagement and support during the testing process. The project’s feasibility study highlights its economic, technical, and social viability, making it a cost-effective and user-centric solution. The system’s architecture, methodology, and modules collectively contribute to a technically sound foundation, ensuring stability and adaptability in various testing scenarios. In this Automated Web Application Testing Platform presented in this paper stands as a valuable tool for addressing the challenges of modern web application testing. With a focus on user-centric design, flexibility, and efficiency, the platform aims to optimize the testing process and contribute to the continuous evolution of software development practices
DSFM Method: A New Approach to Enhancing Discrimination Ability on AI-Generated Datasets
No full textPart 6: AI for ScienceInternational audienceIn recent years, generative large models have achieved remarkable progress, attracting widespread attention. With the rapid development of applications based on these models, public interest in creativity has significantly increased. Generated images have become prevalent in mainstream media and social networks, covering a wide range of topics and domains. Although these technologies have offered unprecedented opportunities for numerous industries, they also come with potential issues such as copyright infringement and information forgery. Existing models for detecting synthetic images typically suffer from low accuracy and weak generalization capabilities. To address these issues, we have proposed a novel method named DSFM. This method utilizes a combination of ResNet and Vision Transformer to simultaneously focus on shallow and deep information, thereby enhancing the overall performance of the model. Experiments conducted on four datasets-AGI, MMAF, Gide COCO, and SFHQ LSUN-demonstrate that our model significantly outperforms baseline models on the MMAF and Gide COCO datasets, with performance improvements nearing 30% on the MMAF dataset. Although improvements on the AGI dataset were modest, the model still displayed competitive performance. Under multiple evaluation metrics, this method has proven its excellent accuracy and superior generalization capabilities. The related datasets and code have been made publicly available and can be accessed at https://github.com/veinhao/ for further information
An Ensemble of Deep Transfer Learning Frameworks for Automatic Tuberculosis Detection in Chest X-Ray Images
No full textPart 2: Applications of AI/ML in Image ProcessingInternational audienceTuberculosis (TB) is a persistent pulmonary disease caused by bacterial infection and is among the ten most prevalent causes of mortality. It is crucial to diagnose TB early since the disease can be fatal if left untreated. Due to technological developments and the availability of medical datasets. An automated system for analysing and classifying chest X-rays (CXR) into TB and non-TB could potentially serve as a dependable substitute for the subjective evaluation performed by medical practitioners. A considerable segment of a CXR image is devoid of diagnostically significant data and is therefore potentially a source of confounding for deep learning (DL) models. CXR image segmentation is performed using the U-Net model; the results of the segmentation are subsequently inputted into the DL models. A classification task was executed on the Montgomery and Shenzhen datasets utilising an ensemble of multiple convolutional neural network (CNN) models, with an assessment of their performance. The proposed ensemble algorithm achieved a higher accuracy of 98.94%. The experimental results show that ensemble learning on segmented lung CXR images produces better results than unsegmented lung CXR images