Dynamic Network Research Articles

HDM-GNN: A Heterogeneous Dynamic Multi-view Graph Neural Network for Crime Prediction

HDM-GNN: A Heterogeneous Dynamic Multi-view Graph Neural Network for Crime Prediction

Energy-Efficient Virtual Network Embedding: A Deep Reinforcement Learning Approach Based on Graph Convolutional Networks

Network virtualization (NV) technology is the cornerstone of modern network architectures, offering significant advantages in resource utilization, flexibility, security, and streamlined management. By enabling the deployment of multiple virtual network requests (VNRs) within a single base network through virtual network embedding (VNE), NV technology can substantially reduce the operational costs and energy consumption. However, the existing algorithms for energy-efficient VNE have limitations, including manual tuning for heuristic routing policies, inefficient feature extraction in traditional intelligent algorithms, and a lack of consideration of periodic traffic fluctuations. To address these limitations, this paper introduces a novel approach that leverages deep reinforcement learning (DRL) to enhance the efficiency of traditional methods. We employ graph convolutional networks (GCNs) for feature extraction, capturing the nuances of network graph structures, and integrate periodic traffic fluctuations as a key constraint in our model. This allows for the predictive embedding of VNRs that is both energy-efficient and responsive to dynamic network conditions. Our research aims to develop an energy-efficient VNE algorithm that dynamically adapts to network traffic patterns, thereby optimizing resource allocation and reducing energy consumption. Extensive simulation experiments demonstrate that our proposed algorithm achieves an average reduction of 22.4% in energy consumption and 41.0% in active substrate nodes, along with a 23.4% improvement in the acceptance rate compared to other algorithms.

The Network Factor of Equity Pricing: A Signed Graph Laplacian Approach

Abstract The connections among firms exhibit heterogeneity, complexity, and dynamism, posing a challenge for traditional unsigned network models. This article proposes a signed graph Laplacian approach to construct a dynamic network index (DNI), quantifying the aggregate changes in the market network over time. A larger DNI indicates more significant changes in firms’ interconnectedness and in the market network structure. Firms with higher sensitivity to DNI exhibit lower expected returns. Incorporating DNI into conventional asset pricing models improves return predictability. Results are robust for multiple estimators, various factor models, and different selections of test assets. Our findings suggest that the network factor generates a significant equity risk premium.

ADDGCN: A Novel Approach with Down-Sampling Dynamic Graph Convolution and Multi-Head Attention for Traffic Flow Forecasting

An essential component of autonomous transportation system management and decision-making is precise and real-time traffic flow forecast. Predicting future traffic conditionsis a difficult undertaking because of the intricate spatio-temporal relationships involved. Existing techniques often employ separate modules to model spatio-temporal features independently, thereby neglecting the temporally and spatially heterogeneous features among nodes. Simultaneously, many existing methods overlook the long-term relationships included in traffic data, subsequently impacting prediction accuracy. We introduce a novel method to traffic flow forecasting based on the combination of the feature-augmented down-sampling dynamic graph convolutional network and multi-head attention mechanism. Our method presents a feature augmentation mechanism to integrate traffic data features at different scales. The subsampled convolutional network enhances information interaction in spatio-temporal data, and the dynamic graph convolutional network utilizes the generated graph structure to better simulate the dynamic relationships between nodes, enhancing the model’s capacity for capturing spatial heterogeneity. Through the feature-enhanced subsampled dynamic graph convolutional network, the model can simultaneously capture spatio-temporal dependencies, and coupled with the process of multi-head temporal attention, it achieves long-term traffic flow forecasting. The findings demonstrate that the ADDGCN model demonstrates superior prediction capabilities on two real datasets (PEMS04 and PEMS08). Notably, for the PEMS04 dataset, compared to the best baseline, the performance of ADDGCN is improved by 2.46% in MAE and 2.90% in RMSE; for the PEMS08 dataset, compared to the best baseline, the ADDGCN performance is improved by 1.50% in RMSE, 3.46% in MAE, and 0.21% in MAPE, indicating our method’s superior performance.

A compositional account of motifs, mechanisms, and dynamics in biochemical regulatory networks

Regulatory networks depict promoting or inhibiting interactions between molecules in a biochemical system. We introduce a category-theoretic formalism for regulatory networks, using signed graphs to model the networks and signed functors to describe occurrences of one network in another, especially occurrences of network motifs. With this foundation, we establish functorial mappings between regulatory networks and other mathematical models in biochemistry. We construct a functor from reaction networks, modeled as Petri nets with signed links, to regulatory networks, enabling us to precisely define when a reaction network could be a physical mechanism underlying a regulatory network. Turning to quantitative models, we associate a regulatory network with a Lotka-Volterra system of differential equations, defining a functor from the category of signed graphs to a category of parameterized dynamical systems. We extend this result from closed to open systems, demonstrating that Lotka-Volterra dynamics respects not only inclusions and collapsings of regulatory networks, but also the process of building up complex regulatory networks by gluing together simpler pieces. Formally, we use the theory of structured cospans to produce a lax double functor from the double category of open signed graphs to that of open parameterized dynamical systems. Throughout the paper, we ground the categorical formalism in examples inspired by systems biology.

Dynamic covalent polymer network enabled bio‐based vitrimers with excellent substrate bonding

AbstractIn this article, we report the synthesis of dynamic covalent polymer network (DCPN)‐enabled vitrimers derived from gum rosin—a pine tree extract. The concept of Diels Alder (DA) thermo‐reversibility has been used to introduce self‐healing/dynamic covalent ability into the polymer systems. For this purpose, the diene moieties were incorporated in the gum rosin‐based polymer network, and the bismaleimides served as the dienophiles. The thermo‐reversible crosslinks formed between the diene and dienophile moieties can enhance the service life of the materials by improving the durability, mechanical as well as chemical properties of the materials. Upon damage, the thermo‐reversible DA bonds de‐crosslink resulting in smaller chain segments, enabling the material to flow to the site of damage and seal the fractured edges. The formation of DA crosslinks has been confirmed by using Fourier transform‐infrared spectroscopic technique. Scanning electron microscopy analysis was used to verify the self‐healing ability of the resin matrix. The healing process was studied after manual scratches were made. The polymers' ability to heal multiple times is investigated using differential scanning calorimetry. The adhesion strength of the self‐healing material has been examined on variety of substrates such as glass, steel, and concrete. The materials exhibited superior adhesion strength on glass substrate.

Social network dynamics, infant loss, and gut microbiota composition in female Colobus vellerosus during time periods with alpha male challenges.

The gut microbiota of group-living animals is strongly influenced by their social interactions, but it is unclear how it responds to social instability. We investigated whether social instability associated with the arrival of new males and challenges to the alpha male position could explain differences in the gut microbiota in adult female Colobus vellerosus at Boabeng-Fiema, Ghana. First, we used a data set collected during May-August 2007 and May 2008-2009 that consisted of (i) 50 fecal samples from adult females in eight social groups for V4 16S rRNA sequencing to determine gut microbiota composition, and (ii) demographic and behavioral data ad libitum to determine male immigration, challenges to the alpha male position, and infant births and deaths. Sørensen and Bray-Curtis beta diversity indices (i.e., between-sample microbiota variation) were predicted by year, alpha male stability, group identity, and age. Next, we used a more detailed behavioral data set collected during focal observations of adult females in one group with a prolonged alpha male takeover and three cases of infant loss, to create 12-month versus 3-month 1-m proximity networks that preceded and overlapped the gut microbiome sampling period in that group. The long versus short-term networks were not correlated, suggesting temporal variation in proximity networks. In this group, beta diversity among the five adult females was predicted by similarity in infant loss status and short-term (rather than yearly) 1-m proximity ties. Although the mechanism driving this association needs to be further investigated in future studies, our findings indicate that alpha male takeovers are associated with gut microbiota variation and highlight the importance of taking demographic and social network dynamics into account.

Prediction of Cleaning Loss of Rice Wheat Combine Harvester Based on Dynamic Bayesian

The cleaning loss rate is a crucial performance metric for rice-wheat combine harvesters. Most studies on the relationship between the cleaning operation parameters of the combined harvester and the cleaning loss rate lack research on the dynamic correlation between harvest losses and operational parameters, specifically focusing on the dynamic correlation between harvest losses and multiple operational parameters such as the cleaning loss rate. In this paper, we formulate a dynamic deductive model for the scenario of harvesting loss regulation involving multiple operational parameters, delving into the dynamic associations between cleaning loss and the regulation of multiple operational parameters and constructing dynamic Bayesian network prediction model. Through experimental analysis of the dynamic Bayesian network prediction model for cleaning loss rates, the comparison of results among the other three algorithms demonstrated that it was efficient for our DBN to predict the cleaning loss rate.

A Construction Method for a Dynamic Weighted Protein Network Using Multi-Level Embedding

The rapid development of high-throughput technology has generated a large amount of protein–protein interaction (PPI) data, which provide a large amount of data support for constructing dynamic protein–protein interaction networks (PPINs). Constructing dynamic PPINs and applying them to recognize protein complexes has become a hot research topic. Most existing methods for complex recognition cannot fully mine the information of PPINs. To address this problem, we propose a construction method of dynamic weighted protein network by multi-level embedding (DWPNMLE). It can reflect the protein network’s dynamics and the protein network’s higher-order proximity. Firstly, the protein active period is calculated to divide the protein subnetworks at different time points. Then, the connection probability is used for the proteins possessing the same time points to judge whether there is an interaction relationship between them. Then, the corresponding protein subnetworks (multiple adjacency matrices) are constructed. Secondly, the multiple feature matrices are constructed using one-hot coding with the gene ontology (GO) information. Next, the first embedding is performed using variational graph auto-encoders (VGAEs) to aggregate features efficiently, followed by the second embedding using deep attributed network embedding (DANE) to strengthen the node representations learned in the first embedding and to maintain the first-order and higher-order proximity of the original network; finally, we compute the cosine similarity to obtain the final dynamic weighted PPIN. To evaluate the effectiveness of DWPNMLE, we apply four classical protein-complex-recognition algorithms on the DWPNMLE and compare them with two other dynamic protein network construction methods. The experimental results demonstrate that DWPNMLE significantly enhances the accuracy of complex recognition with high robustness, and the algorithms’ efficiency is also within a reasonable range.

Molecular Simulation of Covalent Adaptable Networks and Vitrimers: A Review.

Covalent adaptable networks and vitrimers are novel polymers with dynamic reversible bond exchange reactions for crosslinks, enabling them to modulate their properties between those of thermoplastics and thermosets. They have been gathering interest as materials for their recycling and self-healing properties. In this review, we discuss different molecular simulation efforts that have been used over the last decade to investigate and understand the nanoscale and molecular behaviors of covalent adaptable networks and vitrimers. In particular, molecular dynamics, Monte Carlo, and a hybrid of molecular dynamics and Monte Carlo approaches have been used to model the dynamic bond exchange reaction, which is the main mechanism of interest since it controls both the mechanical and rheological behaviors. The molecular simulation techniques presented yield sufficient results to investigate the structure and dynamics as well as the mechanical and rheological responses of such dynamic networks. The benefits of each method have been highlighted. The use of other tools such as theoretical models and machine learning has been included. We noticed, amongst the most prominent results, that stress relaxes as the bond exchange reaction happens, and that at temperatures higher than the glass transition temperature, the self-healing properties are better since more bond BERs are observed. The lifetime of dynamic covalent crosslinks follows, at moderate to high temperatures, an Arrhenius-like temperature dependence. We note the modeling of certain properties like the melt viscosity with glass transition temperature and the topology freezing transition temperature according to a behavior ruled by either the Williams-Landel-Ferry equation or the Arrhenius equation. Discrepancies between the behavior in dissociative and associative covalent adaptable networks are discussed. We conclude by stating which material parameters and atomistic factors, at the nanoscale, have not yet been taken into account and are lacking in the current literature.

DCDIMB: Dynamic Community-based Diversified Influence Maximization using Bridge Nodes

Influence maximization (IM) is the fundamental study of social network analysis. The IM problem finds the top k nodes that have maximum influence in the network. Most of the studies in IM focus on maximizing the number of activated nodes in the static social network. But in real life, social networks are dynamic in nature. This work addresses the diversification of activated nodes in the dynamic social network. This work proposes an objective function that maximizes the number of communities by utilizing bridge nodes. We also propose a diffusion model that considers the role of inactive nodes in influencing a node. We prove the submodularity, and monotonicity of the objective function under the proposed diffusion model. This work analyzes the impact of different ratios of bridge nodes in the seed set on real-world and synthetic datasets. Further, we prove the NP-Hardness of the objective function under the proposed diffusion model. The experiments are conducted on various real-world and synthetic datasets with known and unknown community information. The proposed work experimentally shows that the objective function gives the maximum number of communities considering bridge nodes compared to the benchmark algorithms.

Implemented classification techniques for osteoporosis using deep learning from the perspective of healthcare analytics.

Osteoporosis is a medical disorder that causes bone tissue to deteriorate and lose density, increasing the risk of fractures. Applying Neural Networks (NN) to analyze medical imaging data and detect the presence or severity of osteoporosis in patients is known as osteoporosis classification using Deep Learning (DL) algorithms. DL algorithms can extract relevant information from bone images and discover intricate patterns that could indicate osteoporosis. DCNN biases must be initialized carefully, much like their weights. Biases that are initialized incorrectly might affect the network's learning dynamics and hinder the model's ability to converge to an ideal solution. In this research, Deep Convolutional Neural Networks (DCNNs) are used, which have several benefits over conventional ML techniques for image processing. One of the key benefits of DCNNs is the ability to automatically Feature Extraction (FE) from raw data. Feature learning is a time-consuming procedure in conventional ML algorithms. During the training phase of DCNNs, the network learns to recognize relevant characteristics straight from the data. The Squirrel Search Algorithm (SSA) makes use of a combination of Local Search (LS) and Random Search (RS) techniques that are inspired by the foraging habits of squirrels. The method made it possible to efficiently explore the search space to find prospective values while using promising areas to refine and improve the solutions. Effectively recognizing optimum or nearly optimal solutions depends on balancing exploration and exploitation. The weight in the DCNN is optimized with the help of SSA, which enhances the performance of the classification. The comparative analysis with state-of-the-art techniques shows that the proposed SSA-based DCNN is highly accurate, with 96.57% accuracy.

A lightweight dynamic dual-damped wavelet-based convolutional neural network for interpretable bearing fault diagnosis

Wavelet-based convolutional neural networks (CNNS) have attracted widespread attention because they can improve the interpretability of intelligent fault diagnosis methods. However, the fault feature representation capability of typical wavelet-based convolution kernel frameworks must be strengthened to improve the diagnostic accuracy of complex faults. In the meantime, the large number of network parameters leads to high computational costs. To address these issues, a lightweight wavelet-based dynamic CNN, which comprises a dual-damping wavelet-based dynamic CNN (DWDC) block and a discrete wavelet transformation (DWT) enhancement (DWTE) block, is put forward. In the DWDC block, a wavelet convolution layer is initially designed, where a dual-damping wavelet is used as the kernel function to improve the match of the convolution kernel with fault impulses. Subsequently, a dynamic convolution layer with multiple parallel small-size convolutional kernels is designed to screen the fault features instead of a multilayer network structure, thereby greatly reducing the number of network parameters. Finally, the DWTE block is constructed by combining the DWT and residual dense block, and it can mine more fault information from the previously extracted features. The experiments on the variable speed bearing dataset, locomotive bearing dataset with constant speed and the Case Western Reserve University dataset prove that the proposed approach outperforms five classical CNN models and six advanced wavelet-based CNN models. In addition, it can effectively solve the issue of data imbalance because of its powerful feature extraction capability.

Deep Reinforcement Learning for Optimizing Restricted Access Window in IEEE 802.11ah MAC Layer.

The IEEE 802.11ah standard is introduced to address the growing scale of internet of things (IoT) applications. To reduce contention and enhance energy efficiency in the system, the restricted access window (RAW) mechanism is introduced in the medium access control (MAC) layer to manage the significant number of stations accessing the network. However, to achieve optimized network performance, it is necessary to appropriately determine the RAW parameters, including the number of RAW groups, the number of slots in each RAW, and the duration of each slot. In this paper, we optimize the configuration of RAW parameters in the uplink IEEE 802.11ah-based IoT network. To improve network throughput, we analyze and establish a RAW parameters optimization problem. To effectively cope with the complex and dynamic network conditions, we propose a deep reinforcement learning (DRL) approach to determine the preferable RAW parameters to optimize network throughput. To enhance learning efficiency and stability, we employ the proximal policy optimization (PPO) algorithm. We construct network environments with periodic and random traffic in an NS-3 simulator to validate the performance of the proposed PPO-based RAW parameters optimization algorithm. The simulation results reveal that using the PPO-based DRL algorithm, optimized RAW parameters can be obtained under different network conditions, and network throughput can be improved significantly.

The future of dynamic networks in research and clinical practice.

The future of dynamic networks in research and clinical practice.

Abnormal dynamic functional connectivity in young nondisabling intracerebral hemorrhage patients.

Previous resting-state functional magnetic resonance imaging studies on intracerebral hemorrhage patients have focused more on the static characteristics of brain activity, while the time-varying effects during scanning have received less attention. Therefore, the current study aimed to explore the dynamic functional network connectivity changes of intracerebral hemorrhage patients. Using independent component analysis, the sliding window approach, and the k-means clustering analysis method, different dynamic functional network connectivity states were detected from resting-state functional magnetic resonance imaging data of 37 intracerebral hemorrhage patients and 44 healthy controls. The inter-group differences in dynamic functional network connectivity patterns and temporal properties were investigated, followed by correlation analyses between clinical scales and abnormal functional indexes. Ten resting-state networks were identified, and the dynamic functional network connectivity matrices were clustered into four different states. The transition numbers were decreased in the intracerebral hemorrhage patients compared with healthy controls, which was associated with trail making test scores in patients. The cerebellar network and executive control network connectivity in State 1 was reduced in patients, and this abnormal dynamic functional connectivity was positively correlated with the animal fluency test scores of patients. The current study demonstrated the characteristics of dynamic functional network connectivity in intracerebral hemorrhage patients and revealed that abnormal temporal properties and functional connectivity may be related to the performance of different cognitive domains after ictus. These results may provide new insights into exploring the neurocognitive mechanisms of intracerebral hemorrhage.

Mechanical Fault Diagnosis of High-Voltage Circuit Breakers with Dynamic Multi-Attention Graph Convolutional Networks Based on Adaptive Graph Construction

With the rapid development of deep learning, its powerful capabilities make it possible to perform mechanical fault diagnosis of high-voltage circuit breakers (HVCBs). Among deep learning approaches, the convolutional neural network is widely used. However, while it can extract features effectively, it also has some limitations. Specifically, it depends on a large number of training data and only takes data information into account without considering structural information. These shortcomings lead to unused information and unsatisfactory model results. To address these shortcomings, this paper proposes AKNN-DMGCN, a novel dynamic multi-attention graph convolutional network based on an adaptively constructed graph, which can achieve high accuracy and robust mechanical fault diagnosis of HVCBs. First, a novel adaptive k-nearest neighbor (AKNN) graph construction method is proposed to construct informative graphs. The AKNN method can mine the relationship between the original data samples and utilize the data and label information. Thus, it has high fault tolerance to noise signals and can construct a structure graph with rich and accurate information, which can improve the overall model performance. Then, a dynamic multi-attention graph convolutional network (DMGCN) is applied for mechanical fault diagnosis of HVCBs. DMGCN fully utilizes structural and numerical information representing HVCB signals to perform classification. DMGCN has a dynamic multi-attention mechanism with strong expressive ability, which allows it to achieve high diagnostic accuracy. The experimental results indicate that the accuracy of AKNN-DMGCN reaches 97.22% on a balanced dataset and 95.01% on an imbalanced dataset, which demonstrates that the proposed method is effective for both balanced and imbalanced samples.

Solving the dilemma between reprocessability and mechanical properties faced by recyclable bio-based polyurethane foam through confining imine bonds in the hard domains

Introducing dynamic covalent bonds (DCBs) into polyurethane (PU) foams can solve the problem of their poor recyclability. However, a dilemma arises between reprocessability and mechanical robustness because of the low bond energies of DCBs. In this work, an aromatic diol (VAN-AM) containing imine bonds is synthesized to serve as the polyol together with castor oil. A series of bio-based polyurethane foams (BPUFs) with tunable bio-based carbon content, apparent density, cell properties and mechanical properties are prepared by adjusting the proportions of VAN-AM and castor oil. Benefiting from the molecular rigidity of VAN-AM, the dynamic imine bonds are enriched in the hard domains of PU. This allows for the kinetic control of the dynamicity of the PU networks. When the foams work at ambient temperature, which is much lower than the glass transition temperature of PU (113 ºC), the frozen segments of the hard domains inhibit the dynamic behavior of the imine bonds even if the latter may be somewhat activated (their triggering temperature ≈ 50 ºC). Consequently, the good mechanical properties of BPUFs can be maintained during service. At higher temperatures (> 113 ºC) or swollen state, the segmental mobility in the hard domains and dynamicity of imine bonds are greatly enhanced. Recycling of BPUFs is enabled as evidenced by hot-pressing the crushed foams into the compact sheets under mild conditions. The findings would facilitate the rational design of crosslinked polymeric foams with both high reprocessability and mechanical performance.

Functions of N6-methyladenosine (m6A) RNA modifications in acute myeloid leukemia.

N6-methyladenosine (m6A) is the most common modification of eukaryotic RNA. m6A participates in RNA splicing, nuclear export, translation, and degradation through regulation by methyltransferases, methylation readers, and demethylases, affecting mRNA stability and translation efficiency. Through the dynamic and reversible regulatory network composed of " Writers-Erasers-Readers", m6A modification plays a unique role in the process of hematopoiesis. Acute myeloid leukemia (AML) is a heterogeneous disease characterized by malignant proliferation of hematopoietic stem cells/progenitor cells. Many studies have shown that m6A-related proteins are abnormally expressed in AML and play an important role in the occurrence and development of AML, acting as carcinogenic or anticancer factors. Here, we describe the mechanisms of action of reversing m6A modification in hematopoiesis and AML occurrence and progression to provide a basis for further research on the role of m6A methylation and its regulatory factors in normal hematopoiesis and AML, to ultimately estimate its potential clinical value.

Performance evaluation of the global airline industry under the impact of the COVID-19 pandemic: A dynamic network data envelopment analysis approach

The COVID-19 pandemic posed unprecedented challenges to the airline industry, necessitating a focus on maintaining high efficiency for profitability. This study assesses the efficiency of 26 international airlines from 2019 to 2022 using a dynamic network data envelopment analysis (DNDEA) methodology. The model accounts for the dynamic effect between two consecutive periods and incorporates an internal structure to evaluate airline performance across multiple dimensions. It enables the assessment of overall, period-specific, and stage-specific efficiencies. The findings reveal that while overall efficiency is moderately high on average, no airline achieved full efficiency during the pandemic. Efficiency decreased notably from 2019 to 2020, with a partial recovery but not a return to pre-pandemic levels by 2022. Operational performance remains satisfactory and stable, while service and financial performance exhibit lower efficiency, especially among low-cost airlines compared to full-service counterparts. Additionally, the study explores airlines' environmental impact by considering greenhouse gas emissions. Comparative analysis with a dynamic DEA model without internal structure highlights theoretical contributions, and the study offers managerial insights for airline leaders and policymakers.