43 research outputs found
Accurate prediction of international trade flows: Leveraging knowledge graphs and their embeddings
No full textKnowledge representation (KR) is vital in designing symbolic notations to represent real-world facts and facilitate automated decision-making tasks. Knowledge graphs (KGs) have emerged so far as a popular form of KR, offering a contextual and human-like representation of knowledge. In international economics, KGs have proven valuable in capturing complex interactions between commodities, companies, and countries. By putting the gravity model, which is a common economic framework, into the process of building KGs, important factors that affect trade relationships can be taken into account, making it possible to predict international trade patterns. This paper proposes an approach that leverages Knowledge Graph embeddings for modeling international trade, focusing on link prediction using embeddings. Thus, valuable insights are offered to policymakers, businesses, and economists, enabling them to anticipate the effects of changes in the international trade system. Moreover, the integration of traditional machine learning methods with KG embeddings, such as decision trees and graph neural networks are also explored. The research findings demonstrate the potential for improving prediction accuracy and provide insights into embedding explainability in knowledge representation. The paper also presents a comprehensive analysis of the influence of embedding methods on other intelligent algorithms
Enquête sur l'applicabilité de l'apprentissage en profondeur et de la chaîne de blocs pour l'Internet des objets défini par logiciel
No full text5G mobile network has seen phenomenal growth in providing IoT services and applications. IoT devices are often battery-powered to perform their operations autonomously and serve a variety of situations, such as smart cities, autonomous cars, smart manufacturing, etc., thereby needing efficient energy consumption to extend their lifespan. IoT networks should provide i) an on-demand resource allocation to support adaptive horizontal and vertical scaling of the network resources; ii) flexible infrastructure virtualization that exploits in-network programmability capabilities to operate inside an SDN-enabled virtualization platform; iii) a device-driven and human-driven intelligence to address the issues of energy efficiency and ultra-low latency requirements for future reliable and real-time IoT applications. Despite the promise, IoT networks face several challenging issues stemming from resource constraints and low-computation performance. Additionally, IoT systems encounter several security and privacy concerns to prevent unauthorized access to smart devices and secure trust-less interactions between devices themselves and service providers on the Internet.To address this plethora of challenges, this thesis presents an energy-efficiency IoT system, less computation-intensive, easy to implement, and amenable to online adaptation to the variations of the network condition. In the first contribution, we introduce a novel IoT network virtualization approach based on SDN/NFV to offer a high degree of automation in service chaining delivery for IoT devices. The second contribution introduces a Deep Reinforcement Learning energy-efficient task assignment and scheduling in SDN-based fog IoT Network. Furthermore, we present a computing model for reducing network latency and traffic overhead by centralizing the network control and orchestration in a single SDN controller layer. The last contribution introduces a deep learning approach that combines SDN and blockchain to achieve task scheduling and offloading, improve the response rate of IoT services to offer high levelsof performance, and strive to perform dynamic resource management.Le réseau mobile 5G a connu une croissance phénoménale dans la fourniture de services et d'applications IoT. Les appareils IoT sont souvent alimentés par batterie pour effectuer leurs opérations de manière autonome et servir à diverses situations, telles que les villes intelligentes, les voitures autonomes, la fabrication intelligente, etc. ont donc besoin d'une consommation d'énergie efficace pour prolonger leur durée de vie. Les réseaux IoT devraient fournir: i) une allocation de ressources à la demande pour prendre en charge une mise à l'échelle horizontale et verticale adaptative des ressources du réseau; ii) une virtualisation d'infrastructure flexible qui exploite les capacités de programmabilité en réseau pour fonctionner à l'intérieur d'une plate-forme de virtualisation compatible SDN; iii) une intelligence pilotée par les appareils et pilotée par l'homme pour répondre aux problèmes d'efficacité énergétique et aux exigences de latence ultra-faible pour les futures applications IoT fiables et en temps réel. Malgré la promesse, le réseau IoT est confronté à plusieurs problèmes complexes liés à ses contraintes de ressources et à ses faibles performances de calcul. De plus, les systèmes IoT rencontrent plusieurs problèmes de sécurité et de confidentialité pour empêcher l'accès non autorisé aux appareils intelligents et pour sécuriser les interactions sans confiance entre les appareils eux-mêmes et avec les fournisseurs de services sur Internet.Pour relever cette pléthore de défis, cette thèse présente un système IoT à haut rendement énergétique, moins gourmand en calculs, facile à mettre en œuvre et pouvant être adapté en ligne aux variations de l'état du réseau. Dans la première contribution, nous introduisons une nouvelle approche de virtualisation de réseau IoT basée sur SDN/NFV pour offrir un degré élevé d'automatisation dans la prestation de chaînage de services pour les appareils IoT. Dans la deuxième contribution, nous introduisons une attribution et une planification des tâches économes en énergie par Apprentissage par Renforcement dans un réseau IoT de brouillard basé sur SDN. Nous présentons un modèle informatique pour réduire la latence du réseau et la surcharge de trafic en centralisant le contrôle et l'orchestration du réseau dans une seule couche de contrôleur SDN. La dernière contribution introduit une approche d'apprentissage en profondeur qui combine SDN et blockchain pour réaliser la planification et le déchargement des tâches, améliorer le taux de réponse des services IoT pour offrir des niveaux de performance élevés et s'efforcer d'effectuer une gestion dynamique des ressources
Deep Reinforcement Learning for energy-aware task offloading in join SDN-Blockchain 5G massive IoT edge network
No full textInternational audienceThe Internet-of-Things (IoT) edge allows cloud computing services for topology and location-sensitive distributed computing. As an immediate benefit, it improves network reliability and latency by enabling data access and processing rapidly and efficiently near IoT devices. However, it comes with several issues stemming from the complexity, the security, the energy consumption, and the instability due to the decentralization of service localization. Furthermore, the multi-resource allocation and task scheduling make this task the furthest from being straightforward. Blockchain has been envisioned to enforce trustworthiness in diverse IoT environments. However, high latency and high energy costs are incurred to process IoT transactions. This paper introduces a novel Blockchain-based Deep Reinforcement Learning (DRL) approach to enable energy-aware task scheduling and offloading in an Software Defined Networking (SDN)-enabled IoT network. The Asynchronous Actor-Critic Agent (A3C) DRL-based policy achieves efficient task scheduling and offloading. The latter is in symbiosis with Proof-of-Authority Blockchain consensus to validate IoT transactions and blocks. By doing so, we improve reliability and low latency and achieve energy efficiency for SDNenabled IoT networks. The A3C policy combined with the Blockchain is proved theoretically. Carried out experiments put forth that our approach offers 50% better energy efficiency, which outperforms traditional consensus algorithms, i.e., Proof of Work and PBFT, in terms of throughput and network latency and offers better scheduling performance.</div
Retraction Note: Nickel oxide nanoparticles synthesis using plant extract and evaluation of their antibacterial effects on Streptococcus mutans (Bioprocess and Biosystems Engineering, (2022), 45, 7, (1201-1210), 10.1007/s00449-022-02736-6)
No full textThe Editor-in-Chief has retracted this article because, after publication, concerns were raised regarding the citation of irrelevant references, authorship, and author contributions. The authors were requested to provide explanations for the raised concerns but the Editor-in-Chief found the response provided by the authors insufficient. In the article, references 16, 19, 21–22, 80–81, 108–124 appear to be unrelated to the research described in this article. Additionally, the reference 22 appears to be incorrectly attributed. The Editor-in-Chief therefore no longer has confidence in the reliability of the data presented in this article. Dalal H. Alotaibi, Hanen Sellami, and Mehrdad Khatami do not agree to this retraction. Saade Addalkareem Jasim and Fuad Ameen have not explicitly stated whether they agree to this retraction notice. Nastaran Chokhachi Zadeh Moghadam and Marcos A.L Nobre have not responded to any correspondence from the editor/publisher about this retraction.Department of Pediatric Dentistry Boston University Henry M. Goldman School of Dental MedicineMedical Laboratory Techniques Department Al-Maarif University CollegeDepartment of Botany and Microbiology College of Science King Saud UniversityDepartment of Periodontics and Community Dentistry College of Dentistry King Saud UniversitySchool of Technology and Sciences Sao Paulo State University (Unesp), SPWater Research and Technologies Center (CERTE) Borj-Cedria Technopark University of CarthageAntibacterial Materials R&D Centre China Metal New Materials (Huzhou) Institute, ZhejiangSchool of Technology and Sciences Sao Paulo State University (Unesp), S
Deep Reinforcement Learning for Energy-Efficient Task Scheduling in SDN-based IoT Network
No full textInternational audienceThe growing demand and the diverse traffic patterns coming from various heterogeneous Internet of Things (IoT) systems place an increasing strain on the IoT infrastructure at edge network. Different edge resources (e.g. servers, routers, controllers, gateways) may illustrate different execution times and energy consumption for the same task. They should be capable of achieving high levels of performance to cope with the variability of tasks handling. However, edge nodes are often faced with issues to perform optimal resource distribution and energy-awareness policies in a way that makes effective runtime trade-offs to balance response time constraints, model fidelity, inference accuracy and task schedulability. To address these challenging issues, in this paper we present a dynamic task scheduling and resource management deep reinforcement learning approach for IoT traffic scheduling in SDN-based edge networks. First, we introduce the architectural design of our solution, with the specific objective of achieving high network performance. We formulate a task assignment and scheduling problem that strives to minimize the network latency, while ensuring energy efficiency. The evaluation of our approach offers better results compared against both deterministic and random task scheduling approaches, and show significant performances in terms of latency and energy consumption.</div
Managing Wireless Fog Networks using Software-Defined Networking
No full textInternational audienceFog computing has recently emerged as a new cyber foraging technique to offload resource-intensive tasks from mobile devices to mobile cloudlets in close proximity to end-users. Since the one-hop communication in the network edge is predominantly wireless, Wireless Mesh Networks (WMNs) are being considered to build wireless fog networks. However, WMNs use distributed hop-by-hop routing protocols to reflect a partial visibility of the network, which limits their ability to perform global network management and monitoring needed by fog networks. Software Defined Networking (SDN) provides a centralized control and management of the entire network, which makes it a good candidate to support fog communication. Unfortunately, the SDN OpenFlow protocol does not support any functionalities for wireless fog networks as it is primarily targeted to wired networks. To address these issues, this paper presents a SDN-enabled wireless fog architecture that combines both OpenFlow and distributed wireless protocols. The proposed solution provides lower latency and efficient load balancing to offload the network load by enabling programmable fog routers
Accurate Recommendation of EV Charging Stations Driven by Availability Status Prediction
No full textThe electric vehicle (EV) market is experiencing substantial growth, and it is anticipated to play a major role as a replacement for fossil fuel-powered vehicles in transportation automation systems. Nevertheless, as a rule of thumb, EVs depend on electric charges, where appropriate usage, charging, and energy management are vital requirements. Examining the work that was done before gave us a reason and a basis for making a system that forecasts the real-time availability of electric vehicle charging stations that uses a scalable prediction engine built into a server-side software application that can be used by many people. The implementation process involved scraping data from various sources, creating datasets, and applying feature engineering to the data model. We then applied fundamental models of machine learning to the pre-processed dataset, and subsequently, we proceeded to construct and train an artificial neural network model as the prediction engine. Notably, the results of our research demonstrate that, in terms of precision, recall, and F1-scores, our approach surpasses existing solutions in the literature. These findings underscore the significance of our approach in enhancing the efficiency and usability of EVs, thereby significantly contributing to the acceleration of their adoption in the transportation sector
A SDN-based IoT Architecture Framework forEfficient Energy Management in Smart Buildings
No full textInternational audienceEnergy consumption has increased drastically atglobal scale due to the growing urbanization in cities. Energyefficiency in smart buildings can be achieved by introducing acontext-aware Internet of Things (IoT) approach, where sensorscan learn from their surrounding environment to control theactuators in a coordinated network. However, the IoT networkrequirements are constantly changing in unpredictable fashion,which needs faster and frequent on-demand network reconfigu-ration. Software Defined Network (SDN) has been envisionedas a new approach to enable a flexible and agile networkprogrammability in diverse IoT scenarios. However, the focus hasprimarily been on the design of the SDN computation logic, i.e.controllers, while the dynamic delivery and operations service-inferred IoT resource allocation has been postponed.To fill this gap, this paper proposes a comprehensive architec-tural design that is devised to empower SDN-enabled Context-Aware IoT systems and networks to create efficient energymanagement in smart buildings. We investigate the provisionof NFV IoT functions to support distributed automation andorchestration on IoT devices, and we present a context-awareapproach to gather, filter and process data from sensing data incampus buildings. We provide a proof of concept to demonstratesuccessful deployment and provisioning of virtualized services inthe context of Smart Campus research project
Accurate Energy-Efficient Localization Hybrid Models for Distributed IoT Sensing
No full textInternational audienc
A SDN-based IoT Architecture Framework forEfficient Energy Management in Smart Buildings
No full textInternational audienceEnergy consumption has increased drastically atglobal scale due to the growing urbanization in cities. Energyefficiency in smart buildings can be achieved by introducing acontext-aware Internet of Things (IoT) approach, where sensorscan learn from their surrounding environment to control theactuators in a coordinated network. However, the IoT networkrequirements are constantly changing in unpredictable fashion,which needs faster and frequent on-demand network reconfigu-ration. Software Defined Network (SDN) has been envisionedas a new approach to enable a flexible and agile networkprogrammability in diverse IoT scenarios. However, the focus hasprimarily been on the design of the SDN computation logic, i.e.controllers, while the dynamic delivery and operations service-inferred IoT resource allocation has been postponed.To fill this gap, this paper proposes a comprehensive architec-tural design that is devised to empower SDN-enabled Context-Aware IoT systems and networks to create efficient energymanagement in smart buildings. We investigate the provisionof NFV IoT functions to support distributed automation andorchestration on IoT devices, and we present a context-awareapproach to gather, filter and process data from sensing data incampus buildings. We provide a proof of concept to demonstratesuccessful deployment and provisioning of virtualized services inthe context of Smart Campus research project
