IFIP Open Access Digital Library
Not a member yet
22614 research outputs found
Sort by
Promoting Decarbonization of Islands: A Case Study on the Replacement of Gas Water Heaters in Terceira Island, Azores, Portugal
No full textPart 5: Energy Management and SustainabilityInternational audienceThis study examines the environmental and energy efficiency implications of replacing gas water heaters by electric water heaters (EWHs) in Terceira Island, Azores, Portugal. Through a comprehensive case study, we assess the impact on energy consumption, self-sufficiency, and carbon emissions of the referred replacement. Our findings suggest that transitioning from gas to electric water heating systems can significantly improve energy sustainability metrics, including increased self-consumption and reduced reliance on fossil fuels, therefore contributing to the decarbonization of Terceira Island. This research provides useful insights for policymakers and stakeholders aiming to promote energy sustainability in residential settings
Boosting Digitalization Across European Regions: The AMBITIOUS Approach
No full textPart 1: The 9th Workshop on “5G – Putting Intelligence to the Network Edge” (5G-PINE)International audienceAMBITIOUS [1] aims at providing a fundamental technological infrastructure, which will offer advanced data aggregation and clean-up, analytics, AI-enabled forecasting and secure information exchange mechanisms, via a transparent computing continuum infrastructure, to be integrated with existing, mature services of SMEs, unleashing for them yet unforeseen functionalities and opening up new pathways of commercial exploitation. The envisaged fundamental infrastructure will be provided via the deployment of technological pillars, which will interact with existing services towards supporting the envisioned functionalities. The purpose of a pillar is to provide the same generic functionalities to diverse services, demonstrated by a various set of specific use cases, as a concrete processing chain, aiming at avoiding unnecessary redundancy of resources and budget. This paper presents the focused use cases
Museum Education: Integration of Cultural Heritage and Educational Metadata Schemas
No full textPart 2: The 13th Workshop on “Mining Humanistic Data” (MHDW)International audienceThe emergence of museum education aims at incorporating cultural heritage content within educational metadata frameworks in order to educate versatile individuals, equipped to navigate the complexities of a globalised world. The work navigates through the challenges and benefits of this integration, proposing the development of a unified metadata model for museum education as an incremental step towards improved learning experiences in the digital age. Accordingly, this work proposes the merging of EDM, LOM, and IMS LD metadata schemas so as to enhance the delivery and engagement of educational content by amalgamating it with cultural heritage content. This is achieved through the development a model focused on enhancing learning through CH and its evaluation as far as the opportunities and challenges of this integration through a case study. The case study is centred on the Ionian University Museum’s collections and specifically the “Duplicating Machine Gestetner No 1193069”. The pedagogical benefits of the proposed approach are then showcased, emphasising the enriched learning environment it creates. Thus, the case study demonstrates the feasibility and capability of the model to provide information that will engage students with CH, connecting them with historical and cultural narratives in a meaningful way by integrating detailed cultural and historical insights into the learning process
Open-Source Online Mission-Planning in Emergent Environments with PDDL for Multi-robot Applications
No full textPart 3: The 1st Workshop on “AI Applications for Achieving the Green Deal Targets” (ΑΙ4GD)International audienceThe current study focuses on the development of an open-source framework which is outsourcing the lack of expressivity of the standardized Planning Domain Definition Language – PDDL, leveraging the capacity and flexibility of the hosting environment in Python 3.8. The implementation was based on the definition and logic of a PDDL domain modeling an object stacking/collecting problem, in specified order and locations, by a fleet of autonomous synergetic vehicles by a centralized management agent. A PDDL-parser was utilized to instantiate the solver. Random changes were imposed in the fully-observable environment during mission execution. The proposed framework considered dynamic replanning of the currently valid mission, every time the state space – and consequently the corresponding state graph representation – is stochastically changed. Appropriate test scenarios were defined, to validate the capacity of the implemented framework and establish that it could serve as an interoperable and extensible foundation for other add-ons and graph-searching tools. The evidence from the simulation results indicates the speed and flexibility of the implemented environment for emulating the dynamic replanning problems in highly emergent scenario cases where the state space changes stochastically
Facilitating Gig Work Opportunities for Youth in Developing Countries: A Systematic Literature Review
No full textPart 2: Digital Inclusion Through e-GovernmentInternational audienceGig work presents a promising prospect for youth in developing countries by offering a flexible and economically viable alternative in the face of high unemployment rates. This Systematic Literature Review (SLR) looks to uncover practical strategies and best practices to facilitate the creation of gig work opportunities for youth in developing countries. It poses two research questions, namely: What gig work opportunities exist for youth in a developing country context, and what practices and strategies facilitate the creation of these opportunities? The SLR identifies a number of gig work jobs and groups them into five categories including online consulting, gig work, online micro-tasking, and social media content production. Eleven strategies and two practices for fostering the creation of gig work opportunities in developing countries are identified from the 20 articles included in the study. The strategies range from providing training, improving infrastructure, enacting legislation, raising awareness, to governments creating gig work opportunities by putting tasks on these platforms or creating their own platforms. Consequently, the SLR calls for future research to evaluate the outcomes of strategies and practices implemented to unlock gig work opportunities for youth in developing countries. The SLR also suggests further research into better mapping the exact type of skills necessary to take advantage of the promising gig work opportunities available
An Empirical Analysis of Data Reduction Techniques for k-NN Classification
No full textPart 2: Recommendation/ClassificationInternational audienceThis study explores Data Reduction Techniques (DRTs) in the realm of lazy classification algorithms like k-NN, focusing on Prototype Selection (PS) and Prototype Generation (PG) methods. The research provides an in-depth examination of these methodologies, categorizing DRTs into two primary categories: PS and PG, and further dividing them into three sub-categories: condensation methods, edition methods, and hybrid methods. An experimental study compares a total of 20 new and state-of-the-art DRTs across 20 datasets. The objective is to draw performance conclusions within both the primary and sub-categories, offering valuable insights into how these techniques enhance the effectiveness and robustness of the k-NN classifier. The paper provides a comprehensive overview of DRTs, clarifying their strategies and relative performances
The Faculty Assignment Problem in Higher Education: A Shapley Value-Based Approach
No full textPart 3: Explainable AI - OptimizationInternational audienceThis paper explores the Faculty Assignment Problem in Higher Education with a focus on the Hellenic Open University (HOU), a Distance-Learning Institution. Addressing the challenge of assigning faculty members to teach diverse modules, the paper introduces a Shapley Value-based approach from cooperative game theory. This methodology balances individual faculty expertise across various cognitive subjects, ensuring a more equitable and effective teaching assignment process. It is particularly relevant given the complexity of evaluating over 15,000 candidates for adjunct tutor positions at HOU, where injustices and objections may arise due to the large pool of applicants. The proposed approach, by considering the strategic importance of individual expertise in less represented subjects, aims to enhance the overall quality of education in multidisciplinary and diverse educational programs
A Voting Approach for Explainable Classification with Rule Learning
No full textPart 3: Explainable AI - OptimizationInternational audienceState-of-the-art results in typical classification tasks are mostly achieved by unexplainable machine learning methods, like deep neural networks, for instance. Contrarily, in this paper, we investigate the application of rule learning methods in such a context. Thus, classifications become based on comprehensible (first-order) rules, explaining the predictions made. In general, however, rule-based classifications are less accurate than state-of-the-art results (often significantly). As main contribution, we introduce a voting approach combining both worlds, aiming to achieve comparable results as (unexplainable) state-of-the-art methods, while still providing explanations in the form of comprehensible rules. Considering a variety of benchmark data sets including a use case of significant interest to insurance industries, we prove that our approach not only clearly outperforms ordinary rule learning methods, but also yields results on a par with state-of-the-art outcomes
Optimizations for Learning from Linear Feedback Shift Register Variations with Artificial Neural Networks
No full textPart 3: Explainable AI - OptimizationInternational audienceRecent years have seen an increase in Machine Learning techniques being applied to learn from Pseudorandom Number Generators (PRNGs). Currently, the best results have been obtained for learning on Linear Feedback Shift Registers (LFSRs). Due to the deterministic nature of LFSRs, Decision Trees (DTs) and Artificial Neural Networks (ANNs) were able to reach up to 100% test accuracy for next bit prediction tasks. Despite important advances, a number of directions have been neglected. The current work sets to investigate such directions and bring a more comprehensive understanding of the ANNs capabilities in this context, cementing them as the most reliable technique for learning on LFSRs and their variations. More precisely, the study presents the results of learning from the previously uninvestigated Galois form a LFSR. Moreover, an optimization is proposed with respect to the number of bits needed for learning to predict the outputs of the Geffe generator through the introduction of an ANN pipeline model. Performed experiments display the strength of the approach that is able to maintain up to 100% accuracy for predicting Geffe outputs while reducing the amount of training bits to the degree of magnitude 103 for each LFSR in the proposed pipeline
VulPrompt: Prompt-Based Vulnerability Detection Using Few-Shot Graph Learning
No full textPart 5: ML Attack, VulnerabilityInternational audienceVulPrompt is a new approach for detecting software vulnerabilities from source code by employing a prompt-based graph learning technique within a few-shot learning framework. Rather than adopting the Pretrain-Finetune paradigm typical of prior works, it is the first to adopt the more recent Pretrain-Prompt paradigm in this domain, which affords the creation of a smaller, lightweight model that outperforms larger models within other baseline methods. Evaluations conducted in a few-shot setting reflect the scarcity of large, high-quality labeled datasets for vulnerability detection in large software products—a prevalent issue in cybersecurity. Results show that the reduced number of trainable parameters for prompt-based learning models make them well-suited for this learning scenario, requiring only n instances to train efficiently. The learnable prompt reduces the gap between the pretrain and downstream objectives for a particular task by adjusting the input data for the downstream task to fit the pretrained model. Comparative analyses between VulPrompt and other baseline methods demonstrate the model’s robust performance across all datasets tested, consistently achieving notable results. This success showcases the efficacy and adaptability of VulPrompt for detecting software vulnerabilities across different datasets, highlighting its potential as an impactful tool in the cybersecurity domain