Osmotic energy,existing between the seawater and river water,is a renewable energy source,which can be directly converted into electricity by ion-exchange membranes(IEM).In traditional IEMs,the ion transport channels ...Osmotic energy,existing between the seawater and river water,is a renewable energy source,which can be directly converted into electricity by ion-exchange membranes(IEM).In traditional IEMs,the ion transport channels are formed by nanophase separation of hydrophilic ion carriers and hydrophobic segments.It is difficult to realize high-density ion channels with controlled spatial arrangement and length scale of ion carriers.Herein,we construct high-density 1D ion wires as transmission channels.Through molecular design,hydrophilic imidazole groups and hydrophobic alkyl tails were introduced into the repeat units,which self-assembled into 1D ion transporting core and protecting shell along the main chains.The areal density of the ionic wire arrays is up to~10^(12)cm^(-2),which is the highest value.The ionic wires ensure both high ion flux transport and high selectivity,achieving an ultrahigh-power density of 40.5 W m^(-2)at a 500-fold salinity gradient.Besides,the ionic wire array membrane is well recyclable and antibacterial.The ionic wires provide novel concept for next generation of high-performance membranes.展开更多
Densely deployed Wi Fi networks will play a crucial role in providing the capacity for next generation mobile internet. However, due to increasing interference, overlapped channels in Wi Fi networks and throughput eff...Densely deployed Wi Fi networks will play a crucial role in providing the capacity for next generation mobile internet. However, due to increasing interference, overlapped channels in Wi Fi networks and throughput efficiency degradation, densely deployed Wi Fi networks is not a guarantee to obtain higher throughput. An emergent challenge is how to effi ciently utilize scarce spectrum resources, by matching physical layer resources to traffi c demand. In this aspect, access control allocation strategies play a pivotal role but remain too coarse-grained. As a solution, this research proposes a flexible framework for fine-grained channel width adaptation and multi-channel access in Wi Fi networks. This approach, named SFCA(Subcarrier Fine-grained Channel Access), adopts DOFDM(Discontinuous Orthogonal Frequency Division Multiplexing) at the PHY layer. It allocates the frequency resource with a subcarrier granularity, which facilitates the channel width adaptation for multi-channel access and thus brings more fl exibility and higher frequency efficiency. The MAC layer uses a frequencytime domain backoff scheme, which combines the popular time-domain BEB scheme with a frequency-domain backoff to decrease access collision, resulting in higher access probability for the contending nodes. SFCA is compared with FICA(an established access scheme)showing significant outperformance. Finally we present results for next generation 802.11 ac Wi Fi networks.展开更多
We propose molten polymer's entanglement network deformation to be nonaffine and use transient network structural theory with the revised Liu's kinetics rate equation and the revised upper convected Maxwell co...We propose molten polymer's entanglement network deformation to be nonaffine and use transient network structural theory with the revised Liu's kinetics rate equation and the revised upper convected Maxwell constitutive equation to establish a nonaffine network structural constitutive model for studying the rheological behavior of molten Low Density Polyethylene (LDPE) and High Density Polyethylene (HDPE) in oscillatory shear. As a result, when the strain amplitude or frequency increases, the shear stress amplitude increases. At the same time, the accuracy of the nonaffine network model is higher than that of affine network model. It is clear that there is a small amount of nonaffine network deformation for LDPE melts which have long chain branches, and there is a larger amount of nonaffine network deformation in oscillatory shear for HDPE melts which has no long chain branches. So we had better consider the network deformation nonaffine when we establish the constitutive equations of polymer melts in oscillatory shear.展开更多
In the context of rapid urbanization,high-density construction areas face significant challenges,including the reduction of ecological spaces and the deterioration of their functions.Planning and managing ecological s...In the context of rapid urbanization,high-density construction areas face significant challenges,including the reduction of ecological spaces and the deterioration of their functions.Planning and managing ecological spaces have emerged as essential strategies to address the conflict between urban development and ecological conservation.Using Jinjiang City,Fujian Province as the case study,this paper systematically examines the significance and primary challenges of ecological space planning in highdensity construction areas.It also identifies prevailing issues within the current research domain,including“an overemphasis on top-level design at the expense of implementation,a focus on isolated aspects rather than systemic integration,and prioritization of control over coordination”.This study proposes the key aspects of ecological space planning and management in high-density construction areas,focusing on three fundamental dimensions:human-centered demand orientation,the integration of top-down and bottomup linkage mechanisms,and a differentiated control system.Drawing on the full-element assessment of the ecosystem,ecological network construction,and full-process control system implemented in Jinjiang City,an integrated approach to ecological space governance,encompassing assessment,planning,and control,has been developed.This approach offers both theoretical insights and practical guidance for optimizing ecological spaces in comparable urban contexts.展开更多
Stroke survivors often face significant challenges when performing daily self-care activities due to upper limb motor impairments.Traditional surface electromyography(sEMG)analysis typically focuses on isolated hand p...Stroke survivors often face significant challenges when performing daily self-care activities due to upper limb motor impairments.Traditional surface electromyography(sEMG)analysis typically focuses on isolated hand postures,overlooking the complexity of object-interactive behaviors that are crucial for promoting patient independence.This study introduces a novel framework that combines high-density sEMG(HD-sEMG)signals with an improved Whale Optimization Algorithm(IWOA)-optimized Long Short-Term Memory(LSTM)network to address this limitation.The key contributions of this work include:(1)the creation of a specialized HD-sEMG dataset that captures nine continuous self-care behaviors,along with time and posture markers,to better reflect real-world patient interactions;(2)the development of a multi-channel feature fusion module based on Pascal’s theorem,which enables efficient signal segmentation and spatial–temporal feature extraction;and(3)the enhancement of the IWOA algorithm,which integrates optimal point set initialization,a diversity-driven pooling mechanism,and cosine-based differential evolution to optimize LSTM hyperparameters,thereby improving convergence and global search capabilities.Experimental results demonstrate superior performance,achieving 99.58%accuracy in self-care behavior recognition and 86.19%accuracy for 17 continuous gestures on the Ninapro db2 benchmark.The framework operates with low latency,meeting the real-time requirements for assistive devices.By enabling precise,context-aware recognition of daily activities,this work advances personalized rehabilitation technologies,empowering stroke patients to regain autonomy in self-care tasks.The proposed methodology offers a robust,scalable solution for clinical applications,bridging the gap between laboratory-based gesture recognition and practical,patient-centered care.展开更多
Chemical processes are complex, for which traditional neural network models usually can not lead to satisfactory accuracy. Selective neural network ensemble is an effective way to enhance the generalization accuracy o...Chemical processes are complex, for which traditional neural network models usually can not lead to satisfactory accuracy. Selective neural network ensemble is an effective way to enhance the generalization accuracy of networks, but there are some problems, e.g., lacking of unified definition of diversity among component neural networks and difficult to improve the accuracy by selecting if the diversities of available networks are small. In this study, the output errors of networks are vectorized, the diversity of networks is defined based on the error vectors, and the size of ensemble is analyzed. Then an error vectorization based selective neural network ensemble (EVSNE) is proposed, in which the error vector of each network can offset that of the other networks by training the component networks orderly. Thus the component networks have large diversity. Experiments and comparisons over standard data sets and actual chemical process data set for production of high-density polyethylene demonstrate that EVSNE performs better in generalization ability.展开更多
Metal-support interactions and hydrogen spillover effects in heterogeneous catalysts play a crucial role in aromatic hydrogenation reactions;however,these effects are limited by the metal dispersion on the catalyst an...Metal-support interactions and hydrogen spillover effects in heterogeneous catalysts play a crucial role in aromatic hydrogenation reactions;however,these effects are limited by the metal dispersion on the catalyst and the number of acceptable H*receptors.This study prepares highly dispersed Ni nanoparticles(NPs)catalysts on a Beta substrate via precursor structure topology transformation.In contrast to traditional support materials,the coordination and electronic structure changes between the Ni NPs and the support were achieved,further optimizing the active interface sites and enhancing hydrogen activation and hydrogenation performance.Additionally,the-OH groups at the strong acid sites in zeolite effectively intensified the hydrogen spillover effect as receptors for H^(*)migration and anchoring,accelerating the hydrogenation rate of aromatic rings.Under solvent-free conditions,this catalyst was used for the hydrogenation reaction of aromatic-rich oils,directly producing a C_(8)-C_(14)branched cycloalkanes mixture with an aromatic conversion rate of>99%.The cycloalkanes mixture produced by this method features high density(0.92 g/mL)and a low freezing point(<-60℃),making it suitable for use as high-density aviation fuel or as an additive to enhance the volumetric heat value of conventional aviation fuels in practical applications.展开更多
In rotationally extruded fittings,high-density polyethylene(HDPE)pipes prepared using conventional processing methods often suffer from poor pressure resistance and low toughness.This study introduces an innovative ro...In rotationally extruded fittings,high-density polyethylene(HDPE)pipes prepared using conventional processing methods often suffer from poor pressure resistance and low toughness.This study introduces an innovative rotary shear system(RSS)to address these deficiencies through controlled mandrel rotation and cooling rates.We successfully prepared self-reinforced HDPE pipes with a three-layer structure combining spherical and shish-kebab crystals.Rotational processing aligned the molecular chains in the ring direction and formed shish-kebab crystals.As a result,the annular tensile strength of the rotationally processed three-layer shish-kebab structure(TSK)pipe increased from 26.7 MPa to 76.3 MPa,an enhancement of 185.8%.Notably,while maintaining excellent tensile strength(73.4 MPa),the elongation at break of the spherulite shishkebab spherulite(SKS)tubes was improved to 50.1%,as compared to 33.8%in the case of shish-kebab spherulite shish-kebab(KSK)tubes.This improvement can be attributed to the changes in the micro-morphology and polymer structure within the SKS tubes,specifically due to the formation of small-sized shish-kebab crystals and the low degrees of interlocking.In addition,2D-SAXS analysis revealed that KSK tubes have higher tensile strength due to smaller crystal sizes and larger shish dimensions,forming dense interlocking structures.In contrast,the SKS and TSK tubes had thicker amorphous regions and smaller shish sizes,resulting in reduced interlocking and mechanical performance.展开更多
Evaluation of backfilling effectiveness plays a crucial role in the geological environment management and restoration of abandoned open-pit quarries,providing a scientific basis for subsequent greening efforts.Backfil...Evaluation of backfilling effectiveness plays a crucial role in the geological environment management and restoration of abandoned open-pit quarries,providing a scientific basis for subsequent greening efforts.Backfill soil,predominantly composed of silty clay,demonstrates high water retention capacity and elevated moisture content,leading to a pronounced resistivity contrast with the bedrock exposed by quarrying activities.To investigate the distribution of backfill soil subsurface and assess backfilling effectiveness in the study area,this study conducted a comprehensive geophysical investigation utilizing the high-density electrical resistivity tomography(ERT).A total of 19 ERT survey lines were deployed across three distinct areas in Liuyao Village,Huaibei City,Anhui Province,China.The inversion results,derived from both two-dimensional(2D)and three-dimensional(3D),reveal distinct electrical properties of the subsurface materials:the backfill soil layer shows low resistivity features,the fill stone layer exhibits medium to high resistivity,and the bedrock shows the highest resistivity.The 2D inversion results,from the data measured using the Wenner array effectively capture the spatial distribution and structural features of the backfill soil layer.The findings indicate a gradual east-west thinning of the clay layer within the quarry.Furthermore,the northern pit area exhibits a uniform distribution of backfill soil layer,indicative of effective backfilling operations.In contrast,the southern pit area lacks a well-defined clay layer,suggesting suboptimal backfilling effectiveness.展开更多
High-density interconnect(HDI)soft electronics that can integrate multiple individual functions into one miniaturized monolithic system is promising for applications related to smart healthcare,soft robotics,and human...High-density interconnect(HDI)soft electronics that can integrate multiple individual functions into one miniaturized monolithic system is promising for applications related to smart healthcare,soft robotics,and human-machine interactions.However,despite the recent advances,the development of three-dimensional(3D)soft electronics with both high resolution and high integration is still challenging because of the lack of efficient manufacturing methods to guarantee interlayer alignment of the high-density vias and reliable interlayer electrical conductivity.Here,an advanced 3D laser printing pathway,based on femtosecond laser direct writing(FLDW),is demonstrated for preparing liquid metal(LM)-based any layer HDI soft electronics.FLDW technology,with the characteristics of high spatial resolution and high precision,allows the maskless fabrication of high-resolution embedded LM microchannels and high-density vertical interconnect accesses for 3D integrated circuits.High-aspect-ratio blind/through LM microstructures are formed inside the elastomer due to the supermetalphobicity induced during laser ablation.The LM-based HDI circuit featuring high resolution(~1.5μm)and high integration(10-layer electrical interconnection)is achieved for customized soft electronics,including various customized multilayer passive electric components,soft multilayer circuit,and cross-scale multimode sensors.The 3D laser printing method provides a versatile approach for developing chip-level soft electronics.展开更多
Dislocation strengthening,as one of the methods to simultaneously enhance the room temperature strength and ductility of alloys,does not achieve the desired strengthening and plasticity effect during elevated-temperat...Dislocation strengthening,as one of the methods to simultaneously enhance the room temperature strength and ductility of alloys,does not achieve the desired strengthening and plasticity effect during elevated-temperature deformation.Here,we report a novel strategy to boost the dislocation multiplication and accumulation during deformation at elevated temperatures through dynamic strain aging(DSA).With the introduction of the rare-earth element Ho in Mg-Y-Zn alloy,Ho atoms diffuse toward dislocations during deformation at elevated temperatures,provoking the DSA effect,which increases the dislocation density significantly via the interactions of mobile dislocations and Ho atoms.The resulting alloy achieves a great enhancement of dislocation hardening and obtains the dual benefits of high strength and good ductility simultaneously at high homologous temperatures.The present work provides an effective strategy to enhancing the strength and ductility for elevated-temperature materials.展开更多
BACKGROUND The association between the uric acid-to-high-density lipoprotein cholesterol ratio(UHR)and mental health among individuals with type 2 diabetes mellitus(T2DM)has not been thoroughly investigated.AIM To exa...BACKGROUND The association between the uric acid-to-high-density lipoprotein cholesterol ratio(UHR)and mental health among individuals with type 2 diabetes mellitus(T2DM)has not been thoroughly investigated.AIM To examine the link between UHR and symptoms of depression and anxiety in patients with T2DM.METHODS A cross-sectional analysis was carried out from March 2023 to April 2024,involving participants diagnosed with T2DM.Data on sociodemographic characteristics,clinical parameters,and UHR values were systematically gathered.The Self-Rating Depression Scale(SDS)and Self-Rating Anxiety Scale(SAS)were utilized to evaluate depressive and anxiety symptoms,respectively.To assess the relationships between UHR and SDS/SAS scores,linear regression models were employed,incorporating adjustments for potential confounding variables.Additionally,smooth curve fitting and threshold effect analyses were conducted to explore potential nonlinear relationships.RESULTS A total of 285 patients with T2DM were included.Initial univariate analysis demonstrated a significant positive correlation between elevated UHR levels and higher SDS and SAS scores.Multivariate regression analysis revealed that a one-unit rise in UHR was associated with a 1.13-point increase in SDS scores(95%CI:0.69-1.58)and a 0.57-point increase in SAS scores(95%CI:0.20-0.93).After controlling for confounders,UHR remained positively correlated with SDS(β=1.55,95%CI:0.57-2.53)and SAS(β=0.72,95%CI:0.35-1.09).Nonlinear analysis identified critical thresholds at UHR values of 5.02 for SDS and 4.00 for SAS,beyond which the relationships between UHR and psychological symptom scores became markedly stronger(P<0.05).CONCLUSION Higher UHR levels are significantly linked to exacerbated depressive and anxiety symptoms in patients with T2DM.These results indicate that UHR may function as a promising biomarker to identify individuals at greater risk of mental health complications within this population.展开更多
Rapeseed mustard(Brassica juncea L.) is the third most important oilseed crop in the world, but the geneticmechanism underlying its massive phenotypic variation remains largely unexplored. In this study, specific leng...Rapeseed mustard(Brassica juncea L.) is the third most important oilseed crop in the world, but the geneticmechanism underlying its massive phenotypic variation remains largely unexplored. In this study, specific length amplified fragment sequencing(SLAF-Seq) was used to resequence a population comprising 197 F8recombinantinbred lines(RILs) derived from a cross between vegetable-type Qichi881 and oilseed-type YufengZC of B. juncea. In total, 438,895 high-quality SLAFs were discovered, 47,644 of which were polymorphic, and 3,887 of the polymorphic markers met the requirements for genetic map construction. The final map included 3,887 markers on 18 linkage groups and was 1,830.23 centiMorgan(cM) in length, with an average distance of 0.47 cM between adjacent markers. Using the newly constructed high-density genetic map, a total of 53 QTLs for erucicacid(EA), oleic acid(OA), and linolenic acid(LNA) were detected and integrated into eight consensus QTLswith two for each of these traits. For each of these three traits, two candidate genes were cloned and sequence analysis indicated colocalization with their respective consensus QTLs. The co-dominant allele-specific markers for Bju.FAD3.A03 and Bju.FAD3.B07 were developed and showed co-localization with their consensus QTLs andco-segregation with LNA content, further supporting the results of QTL mapping and bioinformatic analysis. Theexpression levels of the cloned homologous genes were also determined, and the genes were tightly correlatedwith the EA, OA and LNA contents of different lines. The results of this study will facilitate the improvement offatty acid traits and molecular breeding of B. juncea. Further uses of the high-density genetic map created in this study are also discussed.展开更多
Objective Uric acid(UA)to high-density lipoprotein(HDL)ratio(UHR)has recently been proposed as a novel biomarker of inflammation.This study aimed to investigate the association between the UHR and carotid atherosclero...Objective Uric acid(UA)to high-density lipoprotein(HDL)ratio(UHR)has recently been proposed as a novel biomarker of inflammation.This study aimed to investigate the association between the UHR and carotid atherosclerosis(CAS)in patients with type 2 diabetes mellitus(T2DM).Methods In this single-center,retrospective cross-sectional study,379 patients with T2DM were enrolled and categorized into two groups:259 T2DM patients with CAS(T2DM-CAS)and 120 T2DM patients without CAS(T2DM-WCAS).Carotid intima‒media thickness(CIMT)and carotid atheromatous plaques(CAPs)were assessed via Doppler ultrasound.UHR values were compared between the groups,and receiver operating characteristic(ROC)curve analysis was employed to evaluate their diagnostic performance.Results The UHR was significantly greater in the T2DM-CAS group than in the T2DM-WCAS group(P<0.001).Multivariate logistic regression analysis identified the UHR as an independent risk factor for T2DM-CAS(P<0.001).The area under the ROC curve(AUC)for UHR to detect CAS was 0.750,with an optimal cut-off value of 0.35.Conclusion The UHR is an independent risk factor for CAS in patients with T2DM and may serve as a valuable biomarker for predicting CAS in this population.展开更多
Optimizing routing and resource allocation in decentralized unmanned aerial vehicle(UAV)networks remains challenging due to interference and rapidly changing topologies.The authors introduce a novel framework combinin...Optimizing routing and resource allocation in decentralized unmanned aerial vehicle(UAV)networks remains challenging due to interference and rapidly changing topologies.The authors introduce a novel framework combining double deep Q-networks(DDQNs)and graph neural networks(GNNs)for joint routing and resource allocation.The framework uses GNNs to model the network topology and DDQNs to adaptively control routing and resource allocation,addressing interference and improving network performance.Simulation results show that the proposed approach outperforms traditional methods such as Closest-to-Destination(c2Dst),Max-SINR(mSINR),and Multi-Layer Perceptron(MLP)-based models,achieving approximately 23.5% improvement in throughput,50% increase in connection probability,and 17.6% reduction in number of hops,demonstrating its effectiveness in dynamic UAV networks.展开更多
To explore the material basis and mechanisms of the anti-inflammatory effects of Hibiscus mutabilis L..The active ingredients and potential targets of Hibiscus mutabilis L.were obtained through the literature review a...To explore the material basis and mechanisms of the anti-inflammatory effects of Hibiscus mutabilis L..The active ingredients and potential targets of Hibiscus mutabilis L.were obtained through the literature review and SwissADME platform.Genes related to the inflammation were collected using Genecards and OMIM databases,and the intersection genes were submitted on STRING and DAVID websites.Then,the protein interaction network(PPI),gene ontology(GO)and pathway(KEGG)were analyzed.Cytoscape 3.7.2 software was used to construct the“Hibiscus mutabilis L.-active ingredient-target-inflammation”network diagram,and AutoDockTools-1.5.6 software was used for the molecular docking verification.The antiinflammatory effect of Hibiscus mutabilis L.active ingredient was verified by the RAW264.7 inflammatory cell model.The results showed that 11 active components and 94 potential targets,1029 inflammatory targets and 24 intersection targets were obtained from Hibiscus mutabilis L..The key anti-inflammatory active ingredients of Hibiscus mutabilis L.are quercetin,apigenin and luteolin.Its action pathway is mainly related to NF-κB,cancer pathway and TNF signaling pathway.Cell experiments showed that total flavonoids of Hibiscus mutabilis L.could effectively inhibit the expression of tumor necrosis factor(TNF-α),interleukin 8(IL-8)and epidermal growth factor receptor(EGFR)in LPS-induced RAW 264.7 inflammatory cells.It also downregulates the phosphorylation of human nuclear factor ĸB inhibitory protein α(IĸBα)and NF-κB p65 subunit protein(p65).Overall,the anti-inflammatory effect of Hibiscus mutabilis L.is related to many active components,many signal pathways and targets,which provides a theoretical basis for its further development and application.展开更多
Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)t...Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)techniques for DDoS attack diagnosis normally apply network traffic statistical features such as packet sizes and inter-arrival times.However,such techniques sometimes fail to capture complicated relations among various traffic flows.In this paper,we present a new multi-scale ensemble strategy given the Graph Neural Networks(GNNs)for improving DDoS detection.Our technique divides traffic into macro-and micro-level elements,letting various GNN models to get the two corase-scale anomalies and subtle,stealthy attack models.Through modeling network traffic as graph-structured data,GNNs efficiently learn intricate relations among network entities.The proposed ensemble learning algorithm combines the results of several GNNs to improve generalization,robustness,and scalability.Extensive experiments on three benchmark datasets—UNSW-NB15,CICIDS2017,and CICDDoS2019—show that our approach outperforms traditional machine learning and deep learning models in detecting both high-rate and low-rate(stealthy)DDoS attacks,with significant improvements in accuracy and recall.These findings demonstrate the suggested method’s applicability and robustness for real-world implementation in contexts where several DDoS patterns coexist.展开更多
Identifying the community structure of complex networks is crucial to extracting insights and understanding network properties.Although several community detection methods have been proposed,many are unsuitable for so...Identifying the community structure of complex networks is crucial to extracting insights and understanding network properties.Although several community detection methods have been proposed,many are unsuitable for social networks due to significant limitations.Specifically,most approaches depend mainly on user-user structural links while overlooking service-centric,semantic,and multi-attribute drivers of community formation,and they also lack flexible filtering mechanisms for large-scale,service-oriented settings.Our proposed approach,called community discovery-based service(CDBS),leverages user profiles and their interactions with consulted web services.The method introduces a novel similarity measure,global similarity interaction profile(GSIP),which goes beyond typical similarity measures by unifying user and service profiles for all attributes types into a coherent representation,thereby clarifying its novelty and contribution.It applies multiple filtering criteria related to user attributes,accessed services,and interaction patterns.Experimental comparisons against Louvain,Hierarchical Agglomerative Clustering,Label Propagation and Infomap show that CDBS reveals the higher performance as it achieves 0.74 modularity,0.13 conductance,0.77 coverage,and significantly fast response time of 9.8 s,even with 10,000 users and 400 services.Moreover,community discoverybased service consistently detects a larger number of communities with distinct topics of interest,underscoring its capacity to generate detailed and efficient structures in complex networks.These results confirm both the efficiency and effectiveness of the proposed method.Beyond controlled evaluation,communities discovery based service is applicable to targeted recommendations,group-oriented marketing,access control,and service personalization,where communities are shaped not only by user links but also by service engagement.展开更多
With the growing complexity and decentralization of network systems,the attack surface has expanded,which has led to greater concerns over network threats.In this context,artificial intelligence(AI)-based network intr...With the growing complexity and decentralization of network systems,the attack surface has expanded,which has led to greater concerns over network threats.In this context,artificial intelligence(AI)-based network intrusion detection systems(NIDS)have been extensively studied,and recent efforts have shifted toward integrating distributed learning to enable intelligent and scalable detection mechanisms.However,most existing works focus on individual distributed learning frameworks,and there is a lack of systematic evaluations that compare different algorithms under consistent conditions.In this paper,we present a comprehensive evaluation of representative distributed learning frameworks—Federated Learning(FL),Split Learning(SL),hybrid collaborative learning(SFL),and fully distributed learning—in the context of AI-driven NIDS.Using recent benchmark intrusion detection datasets,a unified model backbone,and controlled distributed scenarios,we assess these frameworks across multiple criteria,including detection performance,communication cost,computational efficiency,and convergence behavior.Our findings highlight distinct trade-offs among the distributed learning frameworks,demonstrating that the optimal choice depends strongly on systemconstraints such as bandwidth availability,node resources,and data distribution.This work provides the first holistic analysis of distributed learning approaches for AI-driven NIDS and offers practical guidelines for designing secure and efficient intrusion detection systems in decentralized environments.展开更多
With network attack technology continuing to develop,traditional anomaly traffic detection methods that rely on feature engineering are increasingly insufficient in efficiency and accuracy.Graph Neural Network(GNN),a ...With network attack technology continuing to develop,traditional anomaly traffic detection methods that rely on feature engineering are increasingly insufficient in efficiency and accuracy.Graph Neural Network(GNN),a promising Deep Learning(DL)approach,has proven to be highly effective in identifying intricate patterns in graph⁃structured data and has already found wide applications in the field of network security.In this paper,we propose a hybrid Graph Convolutional Network(GCN)⁃GraphSAGE model for Anomaly Traffic Detection,namely HGS⁃ATD,which aims to improve the accuracy of anomaly traffic detection by leveraging edge feature learning to better capture the relationships between network entities.We validate the HGS⁃ATD model on four publicly available datasets,including NF⁃UNSW⁃NB15⁃v2.The experimental results show that the enhanced hybrid model is 5.71%to 10.25%higher than the baseline model in terms of accuracy,and the F1⁃score is 5.53%to 11.63%higher than the baseline model,proving that the model can effectively distinguish normal traffic from attack traffic and accurately classify various types of attacks.展开更多
基金financially supported by the Key R&D Program of Shandong Province(2022SFGC0801)the National Natural Science Foundation of China(No.22005162 and 22175009)the Natural Science Foundation of Shandong Province(No.ZR2020QE093)。
文摘Osmotic energy,existing between the seawater and river water,is a renewable energy source,which can be directly converted into electricity by ion-exchange membranes(IEM).In traditional IEMs,the ion transport channels are formed by nanophase separation of hydrophilic ion carriers and hydrophobic segments.It is difficult to realize high-density ion channels with controlled spatial arrangement and length scale of ion carriers.Herein,we construct high-density 1D ion wires as transmission channels.Through molecular design,hydrophilic imidazole groups and hydrophobic alkyl tails were introduced into the repeat units,which self-assembled into 1D ion transporting core and protecting shell along the main chains.The areal density of the ionic wire arrays is up to~10^(12)cm^(-2),which is the highest value.The ionic wires ensure both high ion flux transport and high selectivity,achieving an ultrahigh-power density of 40.5 W m^(-2)at a 500-fold salinity gradient.Besides,the ionic wire array membrane is well recyclable and antibacterial.The ionic wires provide novel concept for next generation of high-performance membranes.
基金supported by National Natural Science Foundation of China(No.61471376)the 863 project(No.2014AA01A701)
文摘Densely deployed Wi Fi networks will play a crucial role in providing the capacity for next generation mobile internet. However, due to increasing interference, overlapped channels in Wi Fi networks and throughput efficiency degradation, densely deployed Wi Fi networks is not a guarantee to obtain higher throughput. An emergent challenge is how to effi ciently utilize scarce spectrum resources, by matching physical layer resources to traffi c demand. In this aspect, access control allocation strategies play a pivotal role but remain too coarse-grained. As a solution, this research proposes a flexible framework for fine-grained channel width adaptation and multi-channel access in Wi Fi networks. This approach, named SFCA(Subcarrier Fine-grained Channel Access), adopts DOFDM(Discontinuous Orthogonal Frequency Division Multiplexing) at the PHY layer. It allocates the frequency resource with a subcarrier granularity, which facilitates the channel width adaptation for multi-channel access and thus brings more fl exibility and higher frequency efficiency. The MAC layer uses a frequencytime domain backoff scheme, which combines the popular time-domain BEB scheme with a frequency-domain backoff to decrease access collision, resulting in higher access probability for the contending nodes. SFCA is compared with FICA(an established access scheme)showing significant outperformance. Finally we present results for next generation 802.11 ac Wi Fi networks.
文摘We propose molten polymer's entanglement network deformation to be nonaffine and use transient network structural theory with the revised Liu's kinetics rate equation and the revised upper convected Maxwell constitutive equation to establish a nonaffine network structural constitutive model for studying the rheological behavior of molten Low Density Polyethylene (LDPE) and High Density Polyethylene (HDPE) in oscillatory shear. As a result, when the strain amplitude or frequency increases, the shear stress amplitude increases. At the same time, the accuracy of the nonaffine network model is higher than that of affine network model. It is clear that there is a small amount of nonaffine network deformation for LDPE melts which have long chain branches, and there is a larger amount of nonaffine network deformation in oscillatory shear for HDPE melts which has no long chain branches. So we had better consider the network deformation nonaffine when we establish the constitutive equations of polymer melts in oscillatory shear.
文摘In the context of rapid urbanization,high-density construction areas face significant challenges,including the reduction of ecological spaces and the deterioration of their functions.Planning and managing ecological spaces have emerged as essential strategies to address the conflict between urban development and ecological conservation.Using Jinjiang City,Fujian Province as the case study,this paper systematically examines the significance and primary challenges of ecological space planning in highdensity construction areas.It also identifies prevailing issues within the current research domain,including“an overemphasis on top-level design at the expense of implementation,a focus on isolated aspects rather than systemic integration,and prioritization of control over coordination”.This study proposes the key aspects of ecological space planning and management in high-density construction areas,focusing on three fundamental dimensions:human-centered demand orientation,the integration of top-down and bottomup linkage mechanisms,and a differentiated control system.Drawing on the full-element assessment of the ecosystem,ecological network construction,and full-process control system implemented in Jinjiang City,an integrated approach to ecological space governance,encompassing assessment,planning,and control,has been developed.This approach offers both theoretical insights and practical guidance for optimizing ecological spaces in comparable urban contexts.
基金supported by the National Natural Science Foundation of China(72061006)the research on the auxiliary diagnosis system of chronic injury of levator scapulae based on the concept of digital twin(Contract No:Qian Kehe Support[2023]General 117)Research on indoor intelligent assisted walking robot for the rehabilitation of walking ability of the elderly(Contract No:Qian kehe Support[2023]General 124).
文摘Stroke survivors often face significant challenges when performing daily self-care activities due to upper limb motor impairments.Traditional surface electromyography(sEMG)analysis typically focuses on isolated hand postures,overlooking the complexity of object-interactive behaviors that are crucial for promoting patient independence.This study introduces a novel framework that combines high-density sEMG(HD-sEMG)signals with an improved Whale Optimization Algorithm(IWOA)-optimized Long Short-Term Memory(LSTM)network to address this limitation.The key contributions of this work include:(1)the creation of a specialized HD-sEMG dataset that captures nine continuous self-care behaviors,along with time and posture markers,to better reflect real-world patient interactions;(2)the development of a multi-channel feature fusion module based on Pascal’s theorem,which enables efficient signal segmentation and spatial–temporal feature extraction;and(3)the enhancement of the IWOA algorithm,which integrates optimal point set initialization,a diversity-driven pooling mechanism,and cosine-based differential evolution to optimize LSTM hyperparameters,thereby improving convergence and global search capabilities.Experimental results demonstrate superior performance,achieving 99.58%accuracy in self-care behavior recognition and 86.19%accuracy for 17 continuous gestures on the Ninapro db2 benchmark.The framework operates with low latency,meeting the real-time requirements for assistive devices.By enabling precise,context-aware recognition of daily activities,this work advances personalized rehabilitation technologies,empowering stroke patients to regain autonomy in self-care tasks.The proposed methodology offers a robust,scalable solution for clinical applications,bridging the gap between laboratory-based gesture recognition and practical,patient-centered care.
基金Supported by the National Natural Science Foundation of China (61074153, 61104131)the Fundamental Research Fundsfor Central Universities of China (ZY1111, JD1104)
文摘Chemical processes are complex, for which traditional neural network models usually can not lead to satisfactory accuracy. Selective neural network ensemble is an effective way to enhance the generalization accuracy of networks, but there are some problems, e.g., lacking of unified definition of diversity among component neural networks and difficult to improve the accuracy by selecting if the diversities of available networks are small. In this study, the output errors of networks are vectorized, the diversity of networks is defined based on the error vectors, and the size of ensemble is analyzed. Then an error vectorization based selective neural network ensemble (EVSNE) is proposed, in which the error vector of each network can offset that of the other networks by training the component networks orderly. Thus the component networks have large diversity. Experiments and comparisons over standard data sets and actual chemical process data set for production of high-density polyethylene demonstrate that EVSNE performs better in generalization ability.
基金financially supported by the National Natural Science Foundation of China(Grant 22278439,21776313)the Shandong Province Higher Education Youth Innovation Technology Support Program(Grant 2022KJ074)。
文摘Metal-support interactions and hydrogen spillover effects in heterogeneous catalysts play a crucial role in aromatic hydrogenation reactions;however,these effects are limited by the metal dispersion on the catalyst and the number of acceptable H*receptors.This study prepares highly dispersed Ni nanoparticles(NPs)catalysts on a Beta substrate via precursor structure topology transformation.In contrast to traditional support materials,the coordination and electronic structure changes between the Ni NPs and the support were achieved,further optimizing the active interface sites and enhancing hydrogen activation and hydrogenation performance.Additionally,the-OH groups at the strong acid sites in zeolite effectively intensified the hydrogen spillover effect as receptors for H^(*)migration and anchoring,accelerating the hydrogenation rate of aromatic rings.Under solvent-free conditions,this catalyst was used for the hydrogenation reaction of aromatic-rich oils,directly producing a C_(8)-C_(14)branched cycloalkanes mixture with an aromatic conversion rate of>99%.The cycloalkanes mixture produced by this method features high density(0.92 g/mL)and a low freezing point(<-60℃),making it suitable for use as high-density aviation fuel or as an additive to enhance the volumetric heat value of conventional aviation fuels in practical applications.
基金supported by the National Natural Science Foundation of China(Nos.52373045 and 52033005).
文摘In rotationally extruded fittings,high-density polyethylene(HDPE)pipes prepared using conventional processing methods often suffer from poor pressure resistance and low toughness.This study introduces an innovative rotary shear system(RSS)to address these deficiencies through controlled mandrel rotation and cooling rates.We successfully prepared self-reinforced HDPE pipes with a three-layer structure combining spherical and shish-kebab crystals.Rotational processing aligned the molecular chains in the ring direction and formed shish-kebab crystals.As a result,the annular tensile strength of the rotationally processed three-layer shish-kebab structure(TSK)pipe increased from 26.7 MPa to 76.3 MPa,an enhancement of 185.8%.Notably,while maintaining excellent tensile strength(73.4 MPa),the elongation at break of the spherulite shishkebab spherulite(SKS)tubes was improved to 50.1%,as compared to 33.8%in the case of shish-kebab spherulite shish-kebab(KSK)tubes.This improvement can be attributed to the changes in the micro-morphology and polymer structure within the SKS tubes,specifically due to the formation of small-sized shish-kebab crystals and the low degrees of interlocking.In addition,2D-SAXS analysis revealed that KSK tubes have higher tensile strength due to smaller crystal sizes and larger shish dimensions,forming dense interlocking structures.In contrast,the SKS and TSK tubes had thicker amorphous regions and smaller shish sizes,resulting in reduced interlocking and mechanical performance.
基金Supported by National Key Research and Development Program of China(No.2023YFC3707901)。
文摘Evaluation of backfilling effectiveness plays a crucial role in the geological environment management and restoration of abandoned open-pit quarries,providing a scientific basis for subsequent greening efforts.Backfill soil,predominantly composed of silty clay,demonstrates high water retention capacity and elevated moisture content,leading to a pronounced resistivity contrast with the bedrock exposed by quarrying activities.To investigate the distribution of backfill soil subsurface and assess backfilling effectiveness in the study area,this study conducted a comprehensive geophysical investigation utilizing the high-density electrical resistivity tomography(ERT).A total of 19 ERT survey lines were deployed across three distinct areas in Liuyao Village,Huaibei City,Anhui Province,China.The inversion results,derived from both two-dimensional(2D)and three-dimensional(3D),reveal distinct electrical properties of the subsurface materials:the backfill soil layer shows low resistivity features,the fill stone layer exhibits medium to high resistivity,and the bedrock shows the highest resistivity.The 2D inversion results,from the data measured using the Wenner array effectively capture the spatial distribution and structural features of the backfill soil layer.The findings indicate a gradual east-west thinning of the clay layer within the quarry.Furthermore,the northern pit area exhibits a uniform distribution of backfill soil layer,indicative of effective backfilling operations.In contrast,the southern pit area lacks a well-defined clay layer,suggesting suboptimal backfilling effectiveness.
基金supported by the National Science Foundation of China under the Grant Nos.12127806 and 62175195the International Joint Research Laboratory for Micro/Nano Manufacturing and Measurement Technologies。
文摘High-density interconnect(HDI)soft electronics that can integrate multiple individual functions into one miniaturized monolithic system is promising for applications related to smart healthcare,soft robotics,and human-machine interactions.However,despite the recent advances,the development of three-dimensional(3D)soft electronics with both high resolution and high integration is still challenging because of the lack of efficient manufacturing methods to guarantee interlayer alignment of the high-density vias and reliable interlayer electrical conductivity.Here,an advanced 3D laser printing pathway,based on femtosecond laser direct writing(FLDW),is demonstrated for preparing liquid metal(LM)-based any layer HDI soft electronics.FLDW technology,with the characteristics of high spatial resolution and high precision,allows the maskless fabrication of high-resolution embedded LM microchannels and high-density vertical interconnect accesses for 3D integrated circuits.High-aspect-ratio blind/through LM microstructures are formed inside the elastomer due to the supermetalphobicity induced during laser ablation.The LM-based HDI circuit featuring high resolution(~1.5μm)and high integration(10-layer electrical interconnection)is achieved for customized soft electronics,including various customized multilayer passive electric components,soft multilayer circuit,and cross-scale multimode sensors.The 3D laser printing method provides a versatile approach for developing chip-level soft electronics.
基金supported by the National Key Research and Development Project(2023YFA1609100)the NSFC Funding(U2141207,52171111,52001083)+6 种基金Natural Science Foundation of Heilongjiang(YQ2023E026)China Postdoctoral Science foundation(2024M754149)Postdoctoral Fellowship Program of CPSF(GZC20242192)support from the National Science Foundation(DMR-1611180 and 1809640)with the program directors,DrsHKU Seed Fund for Collaborative Research(#2207101618)support by Croucher Senior Research Fellowship and City U Project(Project No.9229019)Shenzhen Science and Technology Program(Project No.JCYJ20220818101203007)。
文摘Dislocation strengthening,as one of the methods to simultaneously enhance the room temperature strength and ductility of alloys,does not achieve the desired strengthening and plasticity effect during elevated-temperature deformation.Here,we report a novel strategy to boost the dislocation multiplication and accumulation during deformation at elevated temperatures through dynamic strain aging(DSA).With the introduction of the rare-earth element Ho in Mg-Y-Zn alloy,Ho atoms diffuse toward dislocations during deformation at elevated temperatures,provoking the DSA effect,which increases the dislocation density significantly via the interactions of mobile dislocations and Ho atoms.The resulting alloy achieves a great enhancement of dislocation hardening and obtains the dual benefits of high strength and good ductility simultaneously at high homologous temperatures.The present work provides an effective strategy to enhancing the strength and ductility for elevated-temperature materials.
基金Supported by Science and Technology Program of Quzhou,China,No.2022K67Zhejiang Medical Association Clinical Research Fund Project,No.2024ZYC-A526and the Research Project of Quzhou People’s Hospital,No.KYQD2024-006.
文摘BACKGROUND The association between the uric acid-to-high-density lipoprotein cholesterol ratio(UHR)and mental health among individuals with type 2 diabetes mellitus(T2DM)has not been thoroughly investigated.AIM To examine the link between UHR and symptoms of depression and anxiety in patients with T2DM.METHODS A cross-sectional analysis was carried out from March 2023 to April 2024,involving participants diagnosed with T2DM.Data on sociodemographic characteristics,clinical parameters,and UHR values were systematically gathered.The Self-Rating Depression Scale(SDS)and Self-Rating Anxiety Scale(SAS)were utilized to evaluate depressive and anxiety symptoms,respectively.To assess the relationships between UHR and SDS/SAS scores,linear regression models were employed,incorporating adjustments for potential confounding variables.Additionally,smooth curve fitting and threshold effect analyses were conducted to explore potential nonlinear relationships.RESULTS A total of 285 patients with T2DM were included.Initial univariate analysis demonstrated a significant positive correlation between elevated UHR levels and higher SDS and SAS scores.Multivariate regression analysis revealed that a one-unit rise in UHR was associated with a 1.13-point increase in SDS scores(95%CI:0.69-1.58)and a 0.57-point increase in SAS scores(95%CI:0.20-0.93).After controlling for confounders,UHR remained positively correlated with SDS(β=1.55,95%CI:0.57-2.53)and SAS(β=0.72,95%CI:0.35-1.09).Nonlinear analysis identified critical thresholds at UHR values of 5.02 for SDS and 4.00 for SAS,beyond which the relationships between UHR and psychological symptom scores became markedly stronger(P<0.05).CONCLUSION Higher UHR levels are significantly linked to exacerbated depressive and anxiety symptoms in patients with T2DM.These results indicate that UHR may function as a promising biomarker to identify individuals at greater risk of mental health complications within this population.
基金funded by the Scientific and Technological Key Program of Guizhou Province, China (Qiankehezhicheng [2022] Key 031)the National Natural Science Foundation of China (32160483 and 32360497)+2 种基金the Post-Funded Project for the National Natural Science Foundation of China from Guizhou University ([2023]093)the Key Laboratory of Molecular Breeding for Grain and Oil Crops in Guizhou Province, China (Qiankehezhongyindi [2023]008)the Key Laboratory of Functional Agriculture of Guizhou Provincial Higher Education Institutions, China (Qianjiaoji [2023] 007)。
文摘Rapeseed mustard(Brassica juncea L.) is the third most important oilseed crop in the world, but the geneticmechanism underlying its massive phenotypic variation remains largely unexplored. In this study, specific length amplified fragment sequencing(SLAF-Seq) was used to resequence a population comprising 197 F8recombinantinbred lines(RILs) derived from a cross between vegetable-type Qichi881 and oilseed-type YufengZC of B. juncea. In total, 438,895 high-quality SLAFs were discovered, 47,644 of which were polymorphic, and 3,887 of the polymorphic markers met the requirements for genetic map construction. The final map included 3,887 markers on 18 linkage groups and was 1,830.23 centiMorgan(cM) in length, with an average distance of 0.47 cM between adjacent markers. Using the newly constructed high-density genetic map, a total of 53 QTLs for erucicacid(EA), oleic acid(OA), and linolenic acid(LNA) were detected and integrated into eight consensus QTLswith two for each of these traits. For each of these three traits, two candidate genes were cloned and sequence analysis indicated colocalization with their respective consensus QTLs. The co-dominant allele-specific markers for Bju.FAD3.A03 and Bju.FAD3.B07 were developed and showed co-localization with their consensus QTLs andco-segregation with LNA content, further supporting the results of QTL mapping and bioinformatic analysis. Theexpression levels of the cloned homologous genes were also determined, and the genes were tightly correlatedwith the EA, OA and LNA contents of different lines. The results of this study will facilitate the improvement offatty acid traits and molecular breeding of B. juncea. Further uses of the high-density genetic map created in this study are also discussed.
基金supported by the National Natural Science Foundation of China(Grant No.82070862 and 82370840)the National Key R&D Program of China(Grant No.2020YFC2008900).
文摘Objective Uric acid(UA)to high-density lipoprotein(HDL)ratio(UHR)has recently been proposed as a novel biomarker of inflammation.This study aimed to investigate the association between the UHR and carotid atherosclerosis(CAS)in patients with type 2 diabetes mellitus(T2DM).Methods In this single-center,retrospective cross-sectional study,379 patients with T2DM were enrolled and categorized into two groups:259 T2DM patients with CAS(T2DM-CAS)and 120 T2DM patients without CAS(T2DM-WCAS).Carotid intima‒media thickness(CIMT)and carotid atheromatous plaques(CAPs)were assessed via Doppler ultrasound.UHR values were compared between the groups,and receiver operating characteristic(ROC)curve analysis was employed to evaluate their diagnostic performance.Results The UHR was significantly greater in the T2DM-CAS group than in the T2DM-WCAS group(P<0.001).Multivariate logistic regression analysis identified the UHR as an independent risk factor for T2DM-CAS(P<0.001).The area under the ROC curve(AUC)for UHR to detect CAS was 0.750,with an optimal cut-off value of 0.35.Conclusion The UHR is an independent risk factor for CAS in patients with T2DM and may serve as a valuable biomarker for predicting CAS in this population.
文摘Optimizing routing and resource allocation in decentralized unmanned aerial vehicle(UAV)networks remains challenging due to interference and rapidly changing topologies.The authors introduce a novel framework combining double deep Q-networks(DDQNs)and graph neural networks(GNNs)for joint routing and resource allocation.The framework uses GNNs to model the network topology and DDQNs to adaptively control routing and resource allocation,addressing interference and improving network performance.Simulation results show that the proposed approach outperforms traditional methods such as Closest-to-Destination(c2Dst),Max-SINR(mSINR),and Multi-Layer Perceptron(MLP)-based models,achieving approximately 23.5% improvement in throughput,50% increase in connection probability,and 17.6% reduction in number of hops,demonstrating its effectiveness in dynamic UAV networks.
文摘To explore the material basis and mechanisms of the anti-inflammatory effects of Hibiscus mutabilis L..The active ingredients and potential targets of Hibiscus mutabilis L.were obtained through the literature review and SwissADME platform.Genes related to the inflammation were collected using Genecards and OMIM databases,and the intersection genes were submitted on STRING and DAVID websites.Then,the protein interaction network(PPI),gene ontology(GO)and pathway(KEGG)were analyzed.Cytoscape 3.7.2 software was used to construct the“Hibiscus mutabilis L.-active ingredient-target-inflammation”network diagram,and AutoDockTools-1.5.6 software was used for the molecular docking verification.The antiinflammatory effect of Hibiscus mutabilis L.active ingredient was verified by the RAW264.7 inflammatory cell model.The results showed that 11 active components and 94 potential targets,1029 inflammatory targets and 24 intersection targets were obtained from Hibiscus mutabilis L..The key anti-inflammatory active ingredients of Hibiscus mutabilis L.are quercetin,apigenin and luteolin.Its action pathway is mainly related to NF-κB,cancer pathway and TNF signaling pathway.Cell experiments showed that total flavonoids of Hibiscus mutabilis L.could effectively inhibit the expression of tumor necrosis factor(TNF-α),interleukin 8(IL-8)and epidermal growth factor receptor(EGFR)in LPS-induced RAW 264.7 inflammatory cells.It also downregulates the phosphorylation of human nuclear factor ĸB inhibitory protein α(IĸBα)and NF-κB p65 subunit protein(p65).Overall,the anti-inflammatory effect of Hibiscus mutabilis L.is related to many active components,many signal pathways and targets,which provides a theoretical basis for its further development and application.
文摘Distributed Denial of Service(DDoS)attacks are one of the severe threats to network infrastructure,sometimes bypassing traditional diagnosis algorithms because of their evolving complexity.PresentMachine Learning(ML)techniques for DDoS attack diagnosis normally apply network traffic statistical features such as packet sizes and inter-arrival times.However,such techniques sometimes fail to capture complicated relations among various traffic flows.In this paper,we present a new multi-scale ensemble strategy given the Graph Neural Networks(GNNs)for improving DDoS detection.Our technique divides traffic into macro-and micro-level elements,letting various GNN models to get the two corase-scale anomalies and subtle,stealthy attack models.Through modeling network traffic as graph-structured data,GNNs efficiently learn intricate relations among network entities.The proposed ensemble learning algorithm combines the results of several GNNs to improve generalization,robustness,and scalability.Extensive experiments on three benchmark datasets—UNSW-NB15,CICIDS2017,and CICDDoS2019—show that our approach outperforms traditional machine learning and deep learning models in detecting both high-rate and low-rate(stealthy)DDoS attacks,with significant improvements in accuracy and recall.These findings demonstrate the suggested method’s applicability and robustness for real-world implementation in contexts where several DDoS patterns coexist.
文摘Identifying the community structure of complex networks is crucial to extracting insights and understanding network properties.Although several community detection methods have been proposed,many are unsuitable for social networks due to significant limitations.Specifically,most approaches depend mainly on user-user structural links while overlooking service-centric,semantic,and multi-attribute drivers of community formation,and they also lack flexible filtering mechanisms for large-scale,service-oriented settings.Our proposed approach,called community discovery-based service(CDBS),leverages user profiles and their interactions with consulted web services.The method introduces a novel similarity measure,global similarity interaction profile(GSIP),which goes beyond typical similarity measures by unifying user and service profiles for all attributes types into a coherent representation,thereby clarifying its novelty and contribution.It applies multiple filtering criteria related to user attributes,accessed services,and interaction patterns.Experimental comparisons against Louvain,Hierarchical Agglomerative Clustering,Label Propagation and Infomap show that CDBS reveals the higher performance as it achieves 0.74 modularity,0.13 conductance,0.77 coverage,and significantly fast response time of 9.8 s,even with 10,000 users and 400 services.Moreover,community discoverybased service consistently detects a larger number of communities with distinct topics of interest,underscoring its capacity to generate detailed and efficient structures in complex networks.These results confirm both the efficiency and effectiveness of the proposed method.Beyond controlled evaluation,communities discovery based service is applicable to targeted recommendations,group-oriented marketing,access control,and service personalization,where communities are shaped not only by user links but also by service engagement.
基金supported by the Research year project of the KongjuNational University in 2025 and the Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korea government(MSIT)(No.RS-2024-00444170,Research and International Collaboration on Trust Model-Based Intelligent Incident Response Technologies in 6G Open Network Environment).
文摘With the growing complexity and decentralization of network systems,the attack surface has expanded,which has led to greater concerns over network threats.In this context,artificial intelligence(AI)-based network intrusion detection systems(NIDS)have been extensively studied,and recent efforts have shifted toward integrating distributed learning to enable intelligent and scalable detection mechanisms.However,most existing works focus on individual distributed learning frameworks,and there is a lack of systematic evaluations that compare different algorithms under consistent conditions.In this paper,we present a comprehensive evaluation of representative distributed learning frameworks—Federated Learning(FL),Split Learning(SL),hybrid collaborative learning(SFL),and fully distributed learning—in the context of AI-driven NIDS.Using recent benchmark intrusion detection datasets,a unified model backbone,and controlled distributed scenarios,we assess these frameworks across multiple criteria,including detection performance,communication cost,computational efficiency,and convergence behavior.Our findings highlight distinct trade-offs among the distributed learning frameworks,demonstrating that the optimal choice depends strongly on systemconstraints such as bandwidth availability,node resources,and data distribution.This work provides the first holistic analysis of distributed learning approaches for AI-driven NIDS and offers practical guidelines for designing secure and efficient intrusion detection systems in decentralized environments.
基金National Natural Science Foundation of China(Grant No.62103434)National Science Fund for Distinguished Young Scholars(Grant No.62176263).
文摘With network attack technology continuing to develop,traditional anomaly traffic detection methods that rely on feature engineering are increasingly insufficient in efficiency and accuracy.Graph Neural Network(GNN),a promising Deep Learning(DL)approach,has proven to be highly effective in identifying intricate patterns in graph⁃structured data and has already found wide applications in the field of network security.In this paper,we propose a hybrid Graph Convolutional Network(GCN)⁃GraphSAGE model for Anomaly Traffic Detection,namely HGS⁃ATD,which aims to improve the accuracy of anomaly traffic detection by leveraging edge feature learning to better capture the relationships between network entities.We validate the HGS⁃ATD model on four publicly available datasets,including NF⁃UNSW⁃NB15⁃v2.The experimental results show that the enhanced hybrid model is 5.71%to 10.25%higher than the baseline model in terms of accuracy,and the F1⁃score is 5.53%to 11.63%higher than the baseline model,proving that the model can effectively distinguish normal traffic from attack traffic and accurately classify various types of attacks.
