Sign in to use this feature.

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Journals

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Article Types

Countries / Regions

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Search Results (3,343)

Search Parameters:
Keywords = clients

Order results
Result details
Results per page
Select all
Export citation of selected articles as:
10 pages, 427 KiB  
Brief Report
“Big Events” and HIV Transmission Dynamics: Estimating Time Since HIV Infection from Deep Sequencing Data Among Sex Workers and Their Clients in Dnipro, Ukraine
by François Cholette, Nicole Herpai, Leigh M. McClarty, Olga Balakireva, Daria Pavlova, Anna Lopatenko, Rupert Capiña, Paul Sandstrom, Michael Pickles, Evelyn Forget, Sharmistha Mishra, Marissa L. Becker and on behalf of the Dynamics Study Team
Viruses 2025, 17(8), 1148; https://doi.org/10.3390/v17081148 - 21 Aug 2025
Abstract
Background: Major geopolitical events and structural shocks are thought to play a significant role in shaping HIV epidemics by influencing individual behaviours, reshaping social networks, and impacting HIV prevention and treatment programs. Here, we describe individual-level measures of estimated time since HIV infection [...] Read more.
Background: Major geopolitical events and structural shocks are thought to play a significant role in shaping HIV epidemics by influencing individual behaviours, reshaping social networks, and impacting HIV prevention and treatment programs. Here, we describe individual-level measures of estimated time since HIV infection (ETI) from viral next-generation sequencing data among female sex workers and their clients in relation to significant geopolitical events in Ukraine. Methods: The Dynamics Study is a cross-sectional integrated biological and behavioural survey conducted among female sex workers and their clients in Dnipro, Ukraine (December 2017 to March 2018). We were able to successfully sequence a portion of the HIV pol gene on dried blood spot specimens among n = 5/9 clients and n = 5/16 female sex workers who tested positive for HIV (total n = 10/25) using an in-house drug resistance genotyping assay. The “HIV EVO” Intrapatient HIV Evolution web-based tool  was used to infer ETI from viral diversity. Results: The median ETIs for female sex workers and their clients were 5.4 years (IQR = 2.9, 6.6) and 6.5 years (IQR = 5.4, 10.8), respectively. Nearly all HIV acquisition events (n = 7/10; 70%) were estimated to have occurred between the Great Recession (2008–2009) and the War in Donbas (May 2014–February 2022). In general, ETI suggests that HIV acquisition occurred earlier among clients (2012 [IQR = 2007, 2013]) compared to sex workers (2013 [IQR = 2012, 2016]). Conclusion: Our findings suggest that most HIV acquisition in this small subset of female sex workers and clients living with HIV occurred during periods of economic decline. Molecular studies on timing of HIV acquisition against timing of major geopolitical events offer a novel way to contextualize how such events may shape transmission patterns. Full article
(This article belongs to the Section Human Virology and Viral Diseases)
17 pages, 720 KiB  
Article
Neural Network-Based Approaches for Predicting Construction Overruns with Sustainability Considerations
by Kristina Galjanić, Ivan Marović and Tomaš Hanak
Sustainability 2025, 17(16), 7559; https://doi.org/10.3390/su17167559 (registering DOI) - 21 Aug 2025
Abstract
This research focuses on developing neural network-based models for predicting time and cost overruns in construction projects during the construction phase, incorporating sustainability considerations. Previous studies have identified seven key performance areas that affect the final outcome: productivity, quality, time, cost, safety, team [...] Read more.
This research focuses on developing neural network-based models for predicting time and cost overruns in construction projects during the construction phase, incorporating sustainability considerations. Previous studies have identified seven key performance areas that affect the final outcome: productivity, quality, time, cost, safety, team satisfaction, and client satisfaction. Although the interconnections among these performance areas are recognized, their exact relationships and impacts are not fully understood. Hence, the utilization of a neural networks proves to be highly beneficial in predicting the outcome of future construction projects, as it can learn from data and identify patterns, without requiring a complete understanding of these mutual influences. The neural network was trained and tested on the data collected on five completed construction projects, each analyzed at three distinct stages of execution. A total of 182 experiments were conducted to explore different neural network architectures. The most effective configurations for predicting time and cost overruns were identified and evaluated, demonstrating the potential of neural network-based approaches to support more sustainable and proactive project management. The time overrun prediction model demonstrated high accuracy, achieving a MAPE of 10.93%, RMSE of 0.128, and correlation of 0.979. While the cost overrun model showed a lower predictive accuracy, its MAPE (166.76%), RMSE (0.4179), and correlation (0.936) values indicate potential for further refinement. These findings highlight the applicability of neural network-based approaches in construction project management and their potential to support more proactive and informed decision-making. Full article
29 pages, 2673 KiB  
Article
DARTPHROG: A Superscalar Homomorphic Accelerator
by Alexander Magyari and Yuhua Chen
Sensors 2025, 25(16), 5176; https://doi.org/10.3390/s25165176 - 20 Aug 2025
Abstract
Fully Homomorphic Encryption (FHE) allows a client to share their data with an external server without ever exposing their data. FHE serves as a potential solution for data breaches and the marketing of users’ private data. Unfortunately, FHE is much slower than conventional [...] Read more.
Fully Homomorphic Encryption (FHE) allows a client to share their data with an external server without ever exposing their data. FHE serves as a potential solution for data breaches and the marketing of users’ private data. Unfortunately, FHE is much slower than conventional asymmetric cryptography, where data are encrypted only between endpoints. Within this work, we propose the Dynamic AcceleRaTor for Parallel Homomorphic pROGrams, DARTPHROG, as a potential tool for accelerating FHE. DARTPHROG is a superscalar architecture, allowing multiple homomorphic operations to be executed in parallel. Furthermore, DARTPHROG is the first to utilize the new Hardware Optimized Modular-Reduction (HOM-R) system, showcasing the uniquely efficient method compared to Barrett and Montgomery reduction. Coming in at 40.5 W, DARTPHROG is one of the smaller architectures for FHE acceleration. Our architecture offers speedups of up to 1860 times for primitive FHE operations such as ciphertext/plaintext and ciphertext/ciphertext addition, subtraction, and multiplication when operations are performed in parallel using the superscalar feature in DARTPHROG. The DARTPHROG system implements an assembler, a unique instruction set based on THUMB, and a homomorphic processor implemented on a Field Programmable Gate Array (FPGA). DARTPHROG is also the first superscalar evaluation of homomorphic operations when the Number Theoretic Transform (NTT) is excluded from the design. Our processor can therefore be used as a base case for evaluation when weighing the resource and execution impact of NTT implementations. Full article
Show Figures

Figure 1

25 pages, 2133 KiB  
Article
Blockchain-Enabled Self-Autonomous Intelligent Transport System for Drone Task Workflow in Edge Cloud Networks
by Pattaraporn Khuwuthyakorn, Abdullah Lakhan, Arnab Majumdar and Orawit Thinnukool
Algorithms 2025, 18(8), 530; https://doi.org/10.3390/a18080530 - 20 Aug 2025
Abstract
In recent years, self-autonomous intelligent transportation applications such as drones and autonomous vehicles have seen rapid development and deployment across various countries. Within the domain of artificial intelligence, self-autonomous agents are defined as software entities capable of independently operating drones in an intelligent [...] Read more.
In recent years, self-autonomous intelligent transportation applications such as drones and autonomous vehicles have seen rapid development and deployment across various countries. Within the domain of artificial intelligence, self-autonomous agents are defined as software entities capable of independently operating drones in an intelligent transport system (ITS) without human intervention. The integration of these agents into autonomous vehicles and their deployment across distributed cloud networks have increased significantly. These systems, which include drones, ground vehicles, and aircraft, are used to perform a wide range of tasks such as delivering passengers and packages within defined operational boundaries. Despite their growing utility, practical implementations face significant challenges stemming from the heterogeneity of network resources, as well as persistent issues related to security, privacy, and processing costs. To overcome these challenges, this study proposes a novel blockchain-enabled self-autonomous intelligent transport system designed for drone workflow applications. The proposed system architecture is based on a remote method invocation (RMI) client–server model and incorporates a serverless computing framework to manage processing costs. Termed the self-autonomous blockchain-enabled cost-efficient system (SBECES), the framework integrates a client and system agent mechanism governed by Q-learning and deep-learning-based policies. Furthermore, it incorporates a blockchain-based hash validation and fault-tolerant (HVFT) mechanism to ensure data integrity and operational reliability. A deep reinforcement learning (DRL)-enabled adaptive scheduler is utilized to manage drone workflow execution while meeting quality of service (QoS) constraints, including deadlines, cost-efficiency, and security. The overarching objective of this research is to minimize the total processing costs that comprise execution, communication, and security overheads, while maximizing operational rewards and ensuring the timely execution of drone-based tasks. Experimental results demonstrate that the proposed system achieves a 30% reduction in processing costs and a 29% improvement in security and privacy compared to existing state-of-the-art solutions. Full article
(This article belongs to the Section Algorithms for Multidisciplinary Applications)
Show Figures

Figure 1

26 pages, 2266 KiB  
Article
A Phrase Fill-in-Blank Problem in a Client-Side Web Programming Assistant System
by Huiyu Qi, Zhikang Li, Nobuo Funabiki, Htoo Htoo Sandi Kyaw and Wen Chung Kao
Information 2025, 16(8), 709; https://doi.org/10.3390/info16080709 - 20 Aug 2025
Abstract
Mastering client-side Web programming is essential for the development of responsive and interactive Web applications. To support novice students’ self-study, in this paper, we propose a novel exercise format called the phrase fill-in-blank problem (PFP) in the Web Programming Learning Assistant System (WPLAS) [...] Read more.
Mastering client-side Web programming is essential for the development of responsive and interactive Web applications. To support novice students’ self-study, in this paper, we propose a novel exercise format called the phrase fill-in-blank problem (PFP) in the Web Programming Learning Assistant System (WPLAS). A PFP instance presents a source code with blanked phrases (a set of elements) and corresponding Web page screenshots. Then, it requests the user to fill in the blanks, and the answers are automatically evaluated through string matching with predefined correct answers. By increasing blanks, PFP can come close to writing a code from scratch. To facilitate scalable and context-aware question creation, we implemented the PFP instance generation algorithm in Python using regular expressions. This approach targets meaningful code segments in HTML, CSS, and JavaScript that reflect the interactive behavior of front-end development. For evaluations, we generated 10 PFP instances for basic Web programming topics and 5 instances for video games and assigned them to students at Okayama University, Japan, and the State Polytechnic of Malang, Indonesia. Their solution results show that most students could solve them correctly, indicating the effectiveness and accessibility of the generated instances. In addition, we investigated the ability of generative AI, specifically ChatGPT, to solve the PFP instances. The results show 86.7% accuracy for basic-topic PFP instances. Although it still cannot fully find answers, we must monitor progress carefully. In future work, we will enhance PFP in WPLAS to handle non-unique answers by improving answer validation for flexible recognition of equivalent responses. Full article
(This article belongs to the Special Issue Software Applications Programming and Data Security)
Show Figures

Figure 1

13 pages, 2445 KiB  
Article
A Novel Small-Molecule GRP94 Modulator Increases PCSK9 Secretion and Promotes LDLR Degradation
by Wenjing Yan, Yongwang Zhong and Shengyun Fang
Life 2025, 15(8), 1321; https://doi.org/10.3390/life15081321 - 20 Aug 2025
Abstract
The endoplasmic reticulum (ER) maintains protein homeostasis through chaperone-mediated folding and ER-associated degradation (ERAD). Disruption of this quality control, particularly involving the ER chaperone GRP94, contributes to diseases such as hypercholesterolemia, cancer, and immune disorders, where defective GRP94-dependent folding and the trafficking of [...] Read more.
The endoplasmic reticulum (ER) maintains protein homeostasis through chaperone-mediated folding and ER-associated degradation (ERAD). Disruption of this quality control, particularly involving the ER chaperone GRP94, contributes to diseases such as hypercholesterolemia, cancer, and immune disorders, where defective GRP94-dependent folding and the trafficking of client proteins like PCSK9, integrins, and Toll-like receptors drive pathology. Here, we characterize NSC637153 (cp153), a small molecule identified in a drGFP-based ERAD dislocation screen, as a selective probe of GRP94-dependent processes. cp153 inhibits the dislocation of ERAD substrates, preferentially affecting luminal clients, increases PCSK9 secretion, and promotes LDLR degradation. Unlike ATP-competitive HSP90 inhibitors, cp153 does not induce HSP70 or destabilize AKT, suggesting that it perturbs GRP94 function by interfering with client interaction or folding. The identification of cp153 provides a useful tool to for probing GRP94’s role in protein folding, trafficking, ER quality control, and disease-relevant signaling pathways, and supports the development of client-selective GRP94-targeted therapies. Full article
(This article belongs to the Section Physiology and Pathology)
Show Figures

Figure 1

23 pages, 3739 KiB  
Article
FedDPA: Dynamic Prototypical Alignment for Federated Learning with Non-IID Data
by Oussama Akram Bensiah and Rohallah Benaboud
Electronics 2025, 14(16), 3286; https://doi.org/10.3390/electronics14163286 - 19 Aug 2025
Abstract
Federated learning (FL) has emerged as a powerful framework for decentralized model training, preserving data privacy by keeping datasets localized on distributed devices. However, data heterogeneity, characterized by significant variations in size, statistical distribution, and composition across client datasets, presents a persistent challenge [...] Read more.
Federated learning (FL) has emerged as a powerful framework for decentralized model training, preserving data privacy by keeping datasets localized on distributed devices. However, data heterogeneity, characterized by significant variations in size, statistical distribution, and composition across client datasets, presents a persistent challenge that impairs model performance, compromises generalization, and delays convergence. To address these issues, we propose FedDPA, a novel framework that utilizes dynamic prototypical alignment. FedDPA operates in three stages. First, it computes class-specific prototypes for each client to capture local data distributions, integrating them into an adaptive regularization mechanism. Next, a hierarchical aggregation strategy clusters and combines prototypes from similar clients, which reduces communication overhead and stabilizes model updates. Finally, a contrastive alignment process refines the global model by enforcing intra-class compactness and inter-class separation in the feature space. These mechanisms work in concert to mitigate client drift and enhance global model performance. We conducted extensive evaluations on standard classification benchmarks—EMNIST, FEMNIST, CIFAR-10, CIFAR-100, and Tiny-ImageNet 200—under various non-identically and independently distributed (non-IID) scenarios. The results demonstrate the superiority of FedDPA over state-of-the-art methods, including FedAvg, FedNH, and FedROD. Our findings highlight FedDPA’s enhanced effectiveness, stability, and adaptability, establishing it as a scalable and efficient solution to the critical problem of data heterogeneity in federated learning. Full article
Show Figures

Figure 1

34 pages, 639 KiB  
Systematic Review
Federated Learning for Anomaly Detection: A Systematic Review on Scalability, Adaptability, and Benchmarking Framework
by Le-Hang Lim, Lee-Yeng Ong and Meng-Chew Leow
Future Internet 2025, 17(8), 375; https://doi.org/10.3390/fi17080375 - 18 Aug 2025
Viewed by 71
Abstract
Anomaly detection plays an increasingly important role in maintaining the stability and reliability of modern distributed systems. Federated Learning (FL) is an emerging method that shows strong potential in enabling anomaly detection across decentralised environments. However, there are some crucial and tricky research [...] Read more.
Anomaly detection plays an increasingly important role in maintaining the stability and reliability of modern distributed systems. Federated Learning (FL) is an emerging method that shows strong potential in enabling anomaly detection across decentralised environments. However, there are some crucial and tricky research challenges that remain unresolved, such as ensuring scalability, adaptability to dynamic server clusters, and the development of standardised evaluation frameworks for FL. This review aims to address the research gaps through a comprehensive analysis of existing studies. In this paper, a systematic review is conducted by covering three main aspects of the application of FL in anomaly detection: the impact of communication overhead towards scalability and real-time performance, the adaptability of FL frameworks to dynamic server clusters, and the key components required for a standardised benchmarking framework of FL-based anomaly detection. There are a total of 43 relevant articles, published between 2020 and 2025, which were selected from IEEE Xplore, Scopus, and ArXiv. The research findings highlight the potential of asynchronous updates and selective update mechanisms in improving FL’s real-time performance and scalability. This review primarily focuses on anomaly detection tasks in distributed system environments, such as network traffic analysis, IoT devices, and industrial monitoring, rather than domains like computer vision or financial fraud detection. While FL frameworks can handle dynamic client changes, the problem of data heterogeneity among the clients remains a significant obstacle that affects the model convergence speed. Moreover, the lack of a unified benchmarking framework to evaluate the performance of FL in anomaly detection poses a challenge to fair comparisons among the experimental results. Full article
Show Figures

Figure 1

19 pages, 2268 KiB  
Article
Toward the Implementation of Text-Based Web Page Classification and Filtering Solution for Low-Resource Home Routers Using a Machine Learning Approach
by Audronė Janavičiūtė, Agnius Liutkevičius and Nerijus Morkevičius
Electronics 2025, 14(16), 3280; https://doi.org/10.3390/electronics14163280 - 18 Aug 2025
Viewed by 85
Abstract
Restricting and filtering harmful content on the Internet is a serious problem that is often addressed even at the state and legislative levels. Existing solutions for restricting and filtering online content are usually installed on end-user devices and are easily circumvented and difficult [...] Read more.
Restricting and filtering harmful content on the Internet is a serious problem that is often addressed even at the state and legislative levels. Existing solutions for restricting and filtering online content are usually installed on end-user devices and are easily circumvented and difficult to adapt to larger groups of users with different filtering needs. To mitigate this problem, this study proposed a model of a web page classification and filtering solution suitable for use on home routers or other low-resource web page filtering devices. The proposed system combines the constantly updated web page category list approach with machine learning-based text classification methods. Unlike existing web page filtering solutions, such an approach does not require additional software on the client-side, is more difficult to circumvent for ordinary users and can be implemented using common low-resource routers intended for home and organizations usage. This study evaluated the feasibility of the proposed solution by creating the less resource-demanding implementations of machine learning-based web page classification methods adapted for low-resource home routers that could be used to classify and filter unwanted Internet pages in real-time based on the text of the page. The experimental evaluation of softmax regression, decision tree, random forest, and linear SVM (support vector machine) machine learning methods implemented in the C/C++ programming language was performed using a commercial home router Asus RT-AC85P with 256 MB RAM (random access memory) and MediaTek MT7621AT 880 MHz CPU (central processing unit). The implementation of the linear SVM classifier demonstrated the best accuracy of 0.9198 and required 1.86 s to process a web page. The random forest model was only slightly faster (1.56 s to process a web page), while its accuracy reached only 0.7879. Full article
Show Figures

Figure 1

20 pages, 3592 KiB  
Article
Federated Security for Privacy Preservation of Healthcare Data in Edge-Cloud Environments
by Rasanga Jayaweera, Himanshu Agrawal and Nickson M. Karie
Sensors 2025, 25(16), 5108; https://doi.org/10.3390/s25165108 - 17 Aug 2025
Viewed by 279
Abstract
Digital transformation in healthcare has introduced data privacy challenges, as hospitals struggle to protect patient information while adopting digital technologies such as AI, IoT, and cloud more rapidly than ever before. The adoption of powerful third-party Machine Learning as a Service (MLaaS) solutions [...] Read more.
Digital transformation in healthcare has introduced data privacy challenges, as hospitals struggle to protect patient information while adopting digital technologies such as AI, IoT, and cloud more rapidly than ever before. The adoption of powerful third-party Machine Learning as a Service (MLaaS) solutions for disease prediction has become a common practice. However, these solutions offer significant privacy risks when sensitive healthcare data are shared externally to a third-party server. This raises compliance concerns under regulations like HIPAA, GDPR, and Australia’s Privacy Act. To address these challenges, this paper explores a decentralized, privacy-preserving approach to train the models among multiple healthcare stakeholders, integrating Federated Learning (FL) with Homomorphic Encryption (HE), ensuring model parameters remain protected throughout the learning process. This paper proposes a novel Homomorphic Encryption-based Adaptive Tuning for Federated Learning (HEAT-FL) framework to select encryption parameters based on model layer sensitivity. The proposed framework leverages the CKKS scheme to encrypt model parameters on the client side before sharing. This enables secure aggregation at the central server without requiring decryption, providing an additional layer of security through model-layer-wise parameter management. The proposed adaptive encryption approach significantly improves runtime efficiency while maintaining a balanced level of security. Compared to the existing frameworks (non-adaptive) using 256-bit security settings, the proposed framework offers a 56.5% reduction in encryption time for 10 clients and 54.6% for four clients per epoch. Full article
(This article belongs to the Special Issue Privacy and Security in Sensor Networks)
Show Figures

Figure 1

21 pages, 806 KiB  
Tutorial
Multi-Layered Framework for LLM Hallucination Mitigation in High-Stakes Applications: A Tutorial
by Sachin Hiriyanna and Wenbing Zhao
Computers 2025, 14(8), 332; https://doi.org/10.3390/computers14080332 - 16 Aug 2025
Viewed by 386
Abstract
Large language models (LLMs) now match or exceed human performance on many open-ended language tasks, yet they continue to produce fluent but incorrect statements, which is a failure mode widely referred to as hallucination. In low-stakes settings this may be tolerable; in regulated [...] Read more.
Large language models (LLMs) now match or exceed human performance on many open-ended language tasks, yet they continue to produce fluent but incorrect statements, which is a failure mode widely referred to as hallucination. In low-stakes settings this may be tolerable; in regulated or safety-critical domains such as financial services, compliance review, and client decision support, it is not. Motivated by these realities, we develop an integrated mitigation framework that layers complementary controls rather than relying on any single technique. The framework combines structured prompt design, retrieval-augmented generation (RAG) with verifiable evidence sources, and targeted fine-tuning aligned with domain truth constraints. Our interest in this problem is practical. Individual mitigation techniques have matured quickly, yet teams deploying LLMs in production routinely report difficulty stitching them together in a coherent, maintainable pipeline. Decisions about when to ground a response in retrieved data, when to escalate uncertainty, how to capture provenance, and how to evaluate fidelity are often made ad hoc. Drawing on experience from financial technology implementations, where even rare hallucinations can carry material cost, regulatory exposure, or loss of customer trust, we aim to provide clearer guidance in the form of an easy-to-follow tutorial. This paper makes four contributions. First, we introduce a three-layer reference architecture that organizes mitigation activities across input governance, evidence-grounded generation, and post-response verification. Second, we describe a lightweight supervisory agent that manages uncertainty signals and triggers escalation (to humans, alternate models, or constrained workflows) when confidence falls below policy thresholds. Third, we analyze common but under-addressed security surfaces relevant to hallucination mitigation, including prompt injection, retrieval poisoning, and policy evasion attacks. Finally, we outline an implementation playbook for production deployment, including evaluation metrics, operational trade-offs, and lessons learned from early financial-services pilots. Full article
Show Figures

Figure 1

18 pages, 2181 KiB  
Article
MPCTF: A Multi-Party Collaborative Training Framework for Large Language Models
by Ning Liu and Dan Liu
Electronics 2025, 14(16), 3253; https://doi.org/10.3390/electronics14163253 - 16 Aug 2025
Viewed by 248
Abstract
The demand for high-quality private data in large language models is growing significantly. However, private data is often scattered across different entities, leading to significant data silo issues. To alleviate such problems, we propose a novel multi-party collaborative training framework for large language [...] Read more.
The demand for high-quality private data in large language models is growing significantly. However, private data is often scattered across different entities, leading to significant data silo issues. To alleviate such problems, we propose a novel multi-party collaborative training framework for large language models, named MPCTF. MPCTF consists of several components to achieve multi-party collaborative training: (1) a one-click launch mechanism with multi-node and multi-GPU training capabilities, significantly simplifying user operations while enhancing automation and optimizing the collaborative training workflow; (2) four data partitioning strategies for splitting client datasets during the training process, namely fixed-size strategy, percentage-based strategy, maximum data volume strategy, and total data volume and available GPU memory strategy; (3) multiple aggregation strategies; and (4) multiple privacy protection strategies to achieve privacy protection. We conducted extensive experiments to validate the effectiveness of the proposed MPCTF. The experimental results demonstrate that the proposed MPCTF achieves superior performance; for example, our MPCTF acquired an accuracy rate of 65.43 and outperformed the existing work, which acquired an accuracy rate of 14.25 in the experiments. Moreover, we hope that our proposed MPCTF can promote the development of collaborative training for large language models. Full article
(This article belongs to the Special Issue Advances in Information Processing and Network Security)
Show Figures

Figure 1

20 pages, 6757 KiB  
Article
FLUID: Dynamic Model-Agnostic Federated Learning with Pruning and Knowledge Distillation for Maritime Predictive Maintenance
by Alexandros S. Kalafatelis, Angeliki Pitsiakou, Nikolaos Nomikos, Nikolaos Tsoulakos, Theodoros Syriopoulos and Panagiotis Trakadas
J. Mar. Sci. Eng. 2025, 13(8), 1569; https://doi.org/10.3390/jmse13081569 - 15 Aug 2025
Viewed by 267
Abstract
Predictive maintenance (PdM) is vital to maritime operations; however, the traditional deep learning solutions currently offered heavily depend on centralized data aggregation, which is impractical under the limited connectivity, privacy concerns, and resource constraints found in maritime vessels. Federated Learning addresses privacy by [...] Read more.
Predictive maintenance (PdM) is vital to maritime operations; however, the traditional deep learning solutions currently offered heavily depend on centralized data aggregation, which is impractical under the limited connectivity, privacy concerns, and resource constraints found in maritime vessels. Federated Learning addresses privacy by training models locally, yet most FL methods assume homogeneous client architectures and exchange full model weights, leading to heavy communication overhead and sensitivity to system heterogeneity. To overcome these challenges, we introduce FLUID, a dynamic, model-agnostic FL framework that combines client clustering, structured pruning, and student–teacher knowledge distillation. FLUID first groups vessels into resource tiers and calibrates pruning strategies on the most capable client to determine optimal sparsity levels. In subsequent FL rounds, clients exchange logits over a small reference set, decoupling global aggregation from specific model architectures. We evaluate FLUID on a real-world heavy-fuel-oil purifier dataset under realistic heterogeneous deployment. With mixed pruning across clients, FLUID achieves a global R2 of 0.9352, compared with 0.9757 for a centralized baseline. Predictive consistency also remains high for client-based data, with a mean per-client MAE of 0.02575 ± 0.0021 and a mean RMSE of 0.0419 ± 0.0036. These results demonstrate FLUID’s ability to deliver accurate, efficient, and privacy-preserving PdM in heterogeneous maritime fleets. Full article
(This article belongs to the Special Issue Intelligent Solutions for Marine Operations)
Show Figures

Figure 1

20 pages, 2386 KiB  
Article
Personalized Federated Learning Based on Dynamic Parameter Fusion and Prototype Alignment
by Ying Chen, Jing Wen, Shaoling Liang, Zhaofa Chen and Baohua Huang
Sensors 2025, 25(16), 5076; https://doi.org/10.3390/s25165076 - 15 Aug 2025
Viewed by 267
Abstract
To address the limitation of generalization of federated learning under non-independent and identically distributed (Non-IID) data, we propose FedDFPA, a personalized federated learning framework that integrates dynamic parameter fusion and prototype alignment. We design a class-wise dynamic parameter fusion mechanism that adaptively fuses [...] Read more.
To address the limitation of generalization of federated learning under non-independent and identically distributed (Non-IID) data, we propose FedDFPA, a personalized federated learning framework that integrates dynamic parameter fusion and prototype alignment. We design a class-wise dynamic parameter fusion mechanism that adaptively fuses global and local classifier parameters at the class level. It enables each client to preserve its reliable local knowledge while selectively incorporating beneficial global information for personalized classification. We introduce a prototype alignment mechanism based on both global and historical information. By aligning current local features with global prototypes and historical local prototypes, it improves cross-client semantic consistency and enhances the stability of local features. To evaluate the effectiveness of FedDFPA, we conduct extensive experiments on various Non-IID settings and client participation rates. Compared to the average performance of state-of-the-art algorithms, FedDFPA improves the average test accuracy by 3.59% and 4.71% under practical and pathological heterogeneous settings, respectively. These results confirm the effectiveness of our dual-mechanism design in achieving a better balance between personalization and collaboration in federated learning. Full article
(This article belongs to the Section Intelligent Sensors)
Show Figures

Figure 1

51 pages, 1265 KiB  
Review
The Nursing Student Licensure Examination: A Scoping Review
by Flavia Pantaleo, Alessandro Stievano, Chiara Mastroianni, Giorgia Petrucci, Natascia Mazzitelli, Michela Piredda, Maria Grazia De Marinis and Anna Marchetti
Nurs. Rep. 2025, 15(8), 299; https://doi.org/10.3390/nursrep15080299 - 14 Aug 2025
Viewed by 456
Abstract
Background: In an increasingly globalized context marked by growing professional mobility, establishing shared standards for assessing nursing competencies is essential. The licensure examination represents a critical gateway between academic preparation and professional practice. However, significant ambiguity remains regarding what competencies are assessed [...] Read more.
Background: In an increasingly globalized context marked by growing professional mobility, establishing shared standards for assessing nursing competencies is essential. The licensure examination represents a critical gateway between academic preparation and professional practice. However, significant ambiguity remains regarding what competencies are assessed and how this assessment is conducted internationally. Objective: This scoping review aimed to map the international literature on nursing licensure examinations by comparing frameworks and domains, performance levels, and assessment tools to identify similarities and differences in the core competencies required for entry into practice. Methods: The review followed Arksey and O’Malley’s methodological framework. Comprehensive searches were conducted across PubMed, CINAHL, Scopus, ERIC, Cochrane Library, ProQuest, and OpenGrey databases. Studies addressing competency frameworks, performance levels, and assessment tools in undergraduate nursing licensure were included. Results: Twenty-three studies were analyzed. The most frequently cited framework was ‘Client Needs’. Competency domains commonly addressed patient needs, professional roles, and healthcare settings. The dominant performance level was cognitive, typically assessed through multiple-choice questions; practical skills were often evaluated via ‘bedside tests’. Despite variations in frameworks and domains, cognitive performance expectations and assessment tools showed some consistency. Conclusions: These findings underscore the need for a context-sensitive, internationally adaptable framework to promote fairness and support nurse mobility. Full article
(This article belongs to the Section Nursing Education and Leadership)
Show Figures

Graphical abstract

Back to TopTop