Cache management is an important component in any network and it has even more importance in the Future Internet Architectures (FIAs) including Named Data Networking (NDN), because the caches play the key role in redu...
Cache management is an important component in any network and it has even more importance in the Future Internet Architectures (FIAs) including Named Data Networking (NDN), because the caches play the key role in reducing the overall network latency and scalability. In this paper, we discuss the functionality of cache management in NDN, its types as well as its importance for the NDN architecture. In addition, we propose a machine learning-empowered cache management and interests predication for NDN to only preserve the cache only to the secure and really needed data. Our proposal uses Apriori algorithm which is supervised learning algorithm to find the association rules and then to recommend the next requested data. Implementation and experiments on real data traffic depicted that the network’ performance and its influence on the cache increased by 3.2% for two content store sizes of 20 and 40 MB. In addition, a larger cache size of 80 MB shows an increase of the cache hit ratio reaching 90% and hence, clearly reducing the network latency.
In this paper, we propose a model driven approach to facilitate the integration of Nets within Nets (NWN) modeling language in CINCO modeling tool. Despite that, NWN has a sophisticated modeling-simulation tool called...
详细信息
There has been a surge in interest in developing accurate and efficient illness prediction models to help in early diagnosis and treatment planning in recent years. Despite innovations in deep learning techniques, the...
详细信息
Cloud computing, typically provided by large infrastructure providers, allows companies to utilize services efficiently under the premise of cost reduction. Known as the “public cloud,” this model contrasts with the...
详细信息
ISBN:
(数字)9798331527891
ISBN:
(纸本)9798331527907
Cloud computing, typically provided by large infrastructure providers, allows companies to utilize services efficiently under the premise of cost reduction. Known as the “public cloud,” this model contrasts with the “private cloud,” often perceived as costly and complex to implement. However, when high availability is required, the complexities of both models align, and the costs associated with public cloud services can become prohibitive. This paper explores the technical feasibility and economic viability of deploying private cloud infrastructure using refurbished hardware. It presents a detailed architecture incorporating Canonical MaaS for Metal-as-a-Service (MaaS), OpenNebula for Infrastructure-as-a-Service (IaaS), and Incus for Platform-as-a-Service (PaaS), with Moodle as the application layer case study. The study finds that private clouds, especially those utilizing refurbished hardware, not only offer a costeffective alternative to public clouds but also enhance security with complete control over data and systems, provide customizable self-management to meet specific organizational needs, utilize existing resources more efficiently for greater sustainability, and ensure technological sovereignty by reducing dependency on public cloud providers. These benefits collectively underscore the strategic advantages of adopting private cloud infrastructures in certain contexts.
With the increasing focus on healthcare, especially real-time analytics and self-diagnosis, the interest in capturing real-time patient data has increased significantly for both physicians and patients. The developmen...
With the increasing focus on healthcare, especially real-time analytics and self-diagnosis, the interest in capturing real-time patient data has increased significantly for both physicians and patients. The development of smart devices that can collect data such as blood pressure, heart rate, blood sugar level, etc., has greatly contributed to improving individual health management capabilities. However, these advances also, bring challenges such as concerns over safety, reliability, and accurate diagnosis. Problems often arise, especially when exchanging information and transmitting data. To overcome these challenges, this project uses blockchain technology to improve data transmission and security. This study explores the use of blockchain-based internet solutions by integrating blockchain into smart health-care. Additionally, the digital twin concept is used to improve diagnosis by simulating the effects of drugs on virtual objects. Based on the results of these simulations, physicians can make informed decisions about appropriate medications and accurately assess a patient's condition. A health monitoring system using digital twin technology and blockchain can be implemented to securely organize the collection of user information and enable accurate diagnosis. This integration ensures secure storage and transmission of your data and improves your health outcomes. In this paper, we study the integrating blockchain technology with Digital Twin for patient health monitoring. We showcase an interface connected to the blockchain database, specifically designed for doctors to review patient health records through patient wallet address and determine the normalcy of the patient's heart rate based on their medical data. Additionally, the interface allows doctors to predict the effect of medication on the patient's heart rate and blood pressure.
Unmanned aerial vehicles (UAVs) are essential in 5G/6G communication as they provide affordable and effective solutions. However, collecting data in an unfamiliar environment from various sensor nodes is challenging d...
详细信息
Evidence accumulation is thought to be fundamental for decision-making in humans and other mammals. It has been extensively studied in neuroscience and cognitive science with the goal of explaining how sensory informa...
Evidence accumulation is thought to be fundamental for decision-making in humans and other mammals. It has been extensively studied in neuroscience and cognitive science with the goal of explaining how sensory information is sequentially sampled until sufficient evidence has accumulated to favor one decision over others. Neuroscience studies suggest that the hippocampus encodes a low-dimensional ordered representation of evidence through sequential neural activity. Cognitive modelers have proposed a mechanism by which such sequential activity could emerge through the modulation of recurrent weights with a change in the amount of evidence. This gives rise to neurons tuned to a specific magnitude of evidence which resemble neurons recorded in the hippocampus. Here we integrated a cognitive science model inside a Reinforcement Learning (RL) agent and trained the agent to perform a simple evidence accumulation task inspired by the behavioral experiments on animals. We compared the agent's performance with the performance of agents equipped with GRUs and RNNs. We found that the agent based on a cognitive model was able to learn faster and generalize better while having significantly fewer parameters. We also compared the emergent neural activity across agents and found that in some cases, GRU-based agents developed similar neural representations to agents based on a cognitive model. This study illustrates how integrating cognitive models and artificial neural networks can lead to brain-like neural representations that can improve learning.
The rapid proliferation of Unmanned Aerial Vehicles (UAVs) has opened up new horizons in various application domains. However, the optimal selection of UAV services remains a complex challenge, necessitating a nuanced...
详细信息
ISBN:
(数字)9798350350265
ISBN:
(纸本)9798350350272
The rapid proliferation of Unmanned Aerial Vehicles (UAVs) has opened up new horizons in various application domains. However, the optimal selection of UAV services remains a complex challenge, necessitating a nuanced approach considering multiple criteria. This article proposes a novel framework that integrates machine learning (ML) techniques with multi-criteria decision-making (MCDM) methods to address this challenge. Our framework comprehensively evaluates UAV services based on key metrics such as delay, packet loss, throughput, and residual energy, thereby providing a holistic perspective for decision-making. We present experimental results from a comparative study involving Random Forest (RF), Decision Tree (DT), Naive Bayes (NB), and Support Vector Machine (SVM) classifiers. Our findings reveal the effectiveness of our proposed approach, which achieved a classification accuracy of 99.9% with RF. These results underscore the capability of our framework to optimize UAV service delivery, thus enhancing overall performance and user satisfaction. By coupling ML and MCDM, our framework offers a robust and efficient solution for UAV service selection, catering to the diverse needs of users and service providers in the evolving landscape of UAV applications.
Healthcare has undoubtedly brought many advancements through information technology. Specifically, healthcare informatics involves the use of various technologies, data management and communication systems to collect,...
详细信息
Sentiment analysis systems have been shown to exhibit sensitivity to protected attributes. Round-trip translation, on the other hand, has been shown to normalize text. We explore the impact of round-trip translation o...
详细信息
暂无评论