Node Attributes Research Articles

Heterogeneous graph neural network for attribute completion

Heterogeneous graphs consist of multiple types of nodes and edges, and contain comprehensive information and rich semantics, which can properly model real-world complex systems. However, the attribute values of nodes are often incomplete with many missing attributes, as the cost of collecting node attributes is prohibitively expensive or even impossible (e.g., sensitive personal information). While a handful of graph neural network (GNN) models are developed for attribute completion in heterogeneous networks, most of them either ignore the use of similarity between nodes in feature space, or overlook the different importance of different-order neighbor nodes for attribute completion, resulting in poor performance. In this paper, we propose a general Attribute Completion framework for HEterogeneous Networks (AC-HEN), which is composed of feature aggregation, structure aggregation, and multi-view embedding fusion modules. Specifically, AC-HEN leverages feature aggregation and structure aggregation to obtain multi-view embeddings considering neighbor aggregation in both feature space and network structural space, which distinguishes different contributions of different neighbor nodes by conducting weighted aggregation. Then AC-HEN uses the multi-view embeddings to complete the missing attributes via an embedding fusion module in a weak supervised learning paradigm. Extensive experiments on three real-world heterogeneous network datasets demonstrate the superiority of AC-HEN against state-of-the-art baselines in both attribute completion and node classification. The source code is available at: https://github.com/Code-husky/AC-HEN.

Cross-Domain Graph Anomaly Detection.

Anomaly detection on attributed graphs has received increasing research attention lately due to the broad applications in various high-impact domains, such as cybersecurity, finance, and healthcare. Heretofore, most of the existing efforts are predominately performed in an unsupervised manner due to the expensive cost of acquiring anomaly labels, especially for newly formed domains. How to leverage the invaluable auxiliary information from a labeled attributed graph to facilitate the anomaly detection in the unlabeled attributed graph is seldom investigated. In this study, we aim to tackle the problem of cross-domain graph anomaly detection with domain adaptation. However, this task remains nontrivial mainly due to: 1) the data heterogeneity including both the topological structure and nodal attributes in an attributed graph and 2) the complexity of capturing both invariant and specific anomalies on the target domain graph. To tackle these challenges, we propose a novel framework COMMANDER for cross-domain anomaly detection on attributed graphs. Specifically, COMMANDER first compresses the two attributed graphs from different domains to low-dimensional space via a graph attentive encoder. In addition, we utilize a domain discriminator and an anomaly classifier to detect anomalies that appear across networks from different domains. In order to further detect the anomalies that merely appear in the target network, we develop an attribute decoder to provide additional signals for assessing node abnormality. Extensive experiments on various real-world cross-domain graph datasets demonstrate the efficacy of our approach.

Symbolic genetic algorithm for discovering open-form partial differential equations (SGA-PDE)

Partial differential equations (PDEs) are concise and understandable representations of domain knowledge, which are essential for deepening our understanding of physical processes and predicting future responses. However, the PDEs of many real-world problems are uncertain, which calls for PDE discovery. We propose the symbolic genetic algorithm (SGA-PDE) to discover open-form PDEs directly from data without prior knowledge about the equation structure. SGA-PDE focuses on the representation and optimization of PDE. Firstly, SGA-PDE uses symbolic mathematics to realize the flexible representation of any given PDE, transforms a PDE into a forest, and converts each function term into a binary tree. Secondly, SGA-PDE adopts a specially designed genetic algorithm to efficiently optimize the binary trees by iteratively updating the tree topology and node attributes. The SGA-PDE is gradient-free, which is a desirable characteristic in PDE discovery since it is difficult to obtain the gradient between the PDE loss and the PDE structure. In the experiment, SGA-PDE not only successfully discovered nonlinear Burgers' equation, Korteweg-de Vries (KdV) equation, and Chafee-Infante equation, but also handled PDEs with fractional structure and compound functions that cannot be solved by conventional PDE discovery methods.

Network Representation Learning With Community Awareness and Its Applications in Brain Networks.

Previously network representation learning methods mainly focus on exploring the microscopic structure, i.e., the pairwise relationship or similarity between nodes. However, the mesoscopic structure, i.e., community structure, an essential property in real networks, has not been thoroughly studied in the network representation learning. We here propose a deep attributed network representation learning with community awareness (DANRL-CA) framework. Specifically, we design a neighborhood enhancement autoencoder module to capture the 2-step relations between node pairs. To explore the multi-step relations, we construct a community-aware skip-gram module based on the encoder. We introduce two variants of DANRL-CA, namely, DANRL-CA-AM and DANRL-CA-CSM, which incorporate the community information and attribute semantics into node neighbors with different methods. We compare two variant models with the state-of-the-art methods on four datasets for node classification and link prediction. Especially, we apply our models on a brain network. The superiority indicates the scalability and effectiveness of our method on various networks. Compared with DANRL-CA-AM, DANRL-CA-CSM can more flexibly coordinate the role of node attributes and community information in the process of network representation learning, and shows superiority in the networks with sparse topological structure and node attributes.

RKD-VNE: Virtual network embedding algorithm assisted by resource knowledge description and deep reinforcement learning in IIoT scenario

In the era of Industry 4.0, the Industrial Internet of Things (IIoT) is developing rapidly, various IIoT applications pose new challenges to the existing network architecture. On the one hand, these applications put forward higher requirements for the efficient use of network resources. On the other hand, these applications generate massive amounts of information, and they pursue a more secure network environment. Therefore, in order to ensure security while effectively allocating network resources, this paper puts forward a virtual network embedding (VNE) algorithm assisted by resource knowledge description (RKD) and deep reinforcement learning (DRL). First, we use social attribute perception to measure the security of each physical node and regard it as one of the attributes of the physical node. Then, RKD is used to standardize resource constraints before the virtual network is embedded. Finally, the DRL agent derives the probability of the physical node being embedded according to the physical network attributes, and embeds the virtual node according to the probability. Simulation experiments show that compared with the BaseLine algorithm, the RKD-VNE algorithm proposed in this paper has obvious advantages in the general performance of VNE, especially in terms of long-term revenue consumption rate increased by 24.3%.

A Probabilistic Spray-and-Wait Routing Algorithm Based on Node Interest Preference in Delay Tolerant Networks

How to select proper relay nodes to ensure the successful delivery of messages is still a hot topic in delay tolerant networks (DTN). In this paper, we propose a probabilistic Spray-and-Wait routing algorithm based on node interest preference (called NIP-PSW). Firstly, considering the influence of the social attributes of nodes, we define a metric called node interest preference (NIP) to measure the probability of nodes becoming friends. Secondly, in view of the influence of node quality and connection time between nodes on message forwarding, we define the delivery probability (DP). Finally, according to the historical information of nodes, the node interest similarity (NIS) is proposed. In spray phase, NIP and DP are used to select the relay node and allocate the number of message copies adaptively. In wait phase, it is judged whether to forward the message to the encountering nodes again according to the NIS and the DP. In addition, the concept of message storage value (MSV) and the acknowledgment (ACK) mechanism are introduced to manage the buffer of nodes. The simulation results show that the NIP-PSW not only can significantly improve the delivery rate and reduce the average delay but also shows good performance in network overhead and average number of hops.

SpecRp: A spectral-based community embedding algorithm

Community embeddings are useful in node classification since they allow nodes to aggregate relevant information regarding the network structure. Modularity maximization-based algorithms are the most common approach to detect communities in networks. Moreover, spectral theory plays an important role in community detection by providing a matrix representation of the network structure. However, the literature is still scarce on spectral and modularity maximization-based community embedding methods. Besides, the node features of attributed graphs are usually not considered for producing the community embeddings. This paper introduces a community embedding algorithm based on spectral theory, called SpecRp, that has an overlapping modularity maximization-based step also herein proposed. SpecRp is a community detection method that considers node attributes and vertex proximity to obtain community embeddings. Computational experiments showed that SpecRp outperformed the literature in most of the tested benchmark datasets for the node classification task. Moreover, we observed that to detect disjoint communities, SpecRp and the reference literature algorithms presented a conflicting behavior concerning performance measures. While the reference methods achieved better results for modularity, SpecRp performed better concerning the Normalized Mutual Information to the ground-truth partitions. On detecting overlapping communities, SpecRp was considerably faster than the state-of-the-art algorithms, despite presenting worse results in most of the datasets.

Optimal Frameworks for Detecting Anomalies in Sensor-Intensive Heterogeneous Networks

Many network/graph structures are continuously monitored by various sensors that are placed at a subset of nodes and edges. The multidimensional data collected from these sensors over time create large-scale graph data in which the data points are highly dependent. Monitoring large-scale attributed networks with thousands of nodes and heterogeneous sensor data to detect anomalies and unusual events is a complex and computationally expensive process. This paper introduces a new generic approach inspired by state-space models for network anomaly detection that can utilize the information from the network topology, the node attributes (sensor data), and the anomaly propagation sets in an integrated manner to analyze the entire network all at once. This article presents how heterogeneous network sensor data can be analyzed to locate the sources of anomalies as well as the anomalous regions in a network, which can be impacted by one or multiple anomalies at any time instance. Experimental results demonstrate the superior performance of our proposed framework in detecting anomalies in attributed graphs. Summary of Contribution: With the increasing availability of large-scale network sensors and rapid advances in artificial intelligence methods, fundamentally new analytical tools are needed that can integrate data collected from sensors across the networks for decision making while taking into account the stochastic and topological dependencies between nodes, sensors, and anomalies. This paper develops a framework to intelligently and efficiently analyze complex and highly dependent data collected from disparate sensors across large-scale network/graph structures to detect anomalies and abnormal behavior in real time. Unlike general purpose (often black-box) machine learning models, this paper proposes a unique framework for network/graph structures that incorporates the complexities of networks and interdependencies between network entities and sensors. Because of the multidisciplinary nature of the paper that involves optimization, machine learning, and system monitoring and control, it can help researchers in both operations research and computer science domains to develop new network-specific computing tools and machine learning frameworks to efficiently manage large-scale network data.

Embedding graphs on Grassmann manifold

Learning efficient graph representation is the key to favorably addressing downstream tasks on graphs, such as node or graph property prediction. Given the non-Euclidean structural property of graphs, preserving the original graph data’s similarity relationship in the embedded space needs specific tools and a similarity metric. This paper develops a new graph representation learning scheme, namely Egg, which embeds approximated second-order graph characteristics into a Grassmann manifold. The proposed strategy leverages graph convolutions to learn hidden representations of the corresponding subspace of the graph, which is then mapped to a Grassmann point of a low dimensional manifold through truncated singular value decomposition (SVD). The established graph embedding approximates denoised correlationship of node attributes, as implemented in the form of a symmetric matrix space for Euclidean calculation. The effectiveness of Egg is demonstrated using both clustering and classification tasks at the node level and graph level. It outperforms baseline models on various benchmarks.

Variational Graph Autoencoder with Mutual Information Maximization for Graph Representations Learning

Graph neural network (GNN) is a powerful representation learning framework for graph-structured data. Some GNN-based graph embedding methods, including variational graph autoencoder (VGAE), have been presented recently. However, existing VGAE-based methods typically focus on reconstructing the adjacent matrix, i.e. topological structure, instead of the node features matrix, this strategy makes graphical features difficult to be fully learned, which weakens and restricts the capacity of a generative network to learn higher-quality representations. To address the issue, we use a contrastive estimator on the representation mechanism, i.e. on the encoding process under the framework of VGAE. In particular, we maximize the mutual information (MI) between encoded latent representation and node attributes which acts as a regularizer forcing the encoder to select the most informative with respect to the node attributes. Additionally, we also solve another key question how to effectively estimate the mutual information by drawing samples from the joint and marginal, and explain why the maximization of MI can contribute to the encoder obtaining more node feature information. Ultimately, extensive experiments on three citation networks and four web-age networks show that our method outperforms contemporary popular algorithms (such as DGI) on node classifications and clustering tasks, and the best result is an [Formula: see text] increase on node clustering task.

Measuring and sampling: A metric‐guided subgraph learning framework for graph neural network

Graph neural networks (GNNs) have shown convincing performance in learning powerful node representations that preserve both node attributes and graph structural information. However, many GNNs encounter problems in effectiveness and efficiency when they are designed with a deeper network structure or handle large-sized graphs. Several sampling algorithms have been proposed for improving and accelerating the training of GNNs, yet they ignore understanding the source of GNNs performance gain. The measurement of information within graph data can help the sampling algorithms to keep high-value information while removing redundant information and even noise. In this paper, we propose a Metric-Guided (MeGuide) subgraph learning framework for GNNs. MeGuide employs two novel metrics: Feature Smoothness and Connection Failure Distance to guide the subgraph sampling and mini-batch based training. Feature Smoothness is designed for analyzing the feature of nodes to retain the most valuable information, while Connection Failure Distance can measure the structural information to control the size of subgraphs. We demonstrate the effectiveness and efficiency of MeGuide in training various GNNs on multiple data sets.

An Efficient Routing Scheme Based on Node Attributes for Opportunistic Networks in Oceans.

Along with the fast development of the marine economy and ever-increasing human activities, handy and reliable marine networking services are increasingly required in recent years. The ocean faces challenges to support cost-effective communication due to its special environments. Opportunistic networks with easy deployment and self-curing capability are expected to play an important role to adapt to such dynamic networking environments. In the literature, routing schemes for opportunistic networks mainly exploit node mobility and local relaying technologies. They did not take into account the impact of node behaviors on encountering opportunities and in case of no further relaying, network performance would be greatly degraded. To solve the problem, we propose an efficient routing scheme based on node attributes for opportunistic networks. We first construct delivery competency to predict the further relay nodes. Then a forwarding willingness mechanism is introduced to evaluate the relaying probability combining device capacity and movement behaviors of nodes. Finally, the utility metric is used to make decisions on message forwarding. The results show that the proposed scheme improves network performance in terms of delivery ratio, average latency, and overhead ratio as compared to other schemes.

White Matter Network Disruption Is Associated With Melancholic Features in Major Depressive Disorder.

BackgroundThe efficacy and prognosis of major depressive disorder (MDD) are limited by its heterogeneity. MDD with melancholic features is an important subtype of MDD. The present study aimed to reveal the white matter (WM) network changes in melancholic depression.Materials and MethodsTwenty-three first-onset, untreated melancholic MDD, 59 non-melancholic MDD patients and 63 health controls underwent diffusion tensor imaging (DTI) scans. WM network analysis based on graph theory and support vector machine (SVM) were used for image data analysis.ResultsCompared with HC, small-worldness was reduced and abnormal node attributes were in the right orbital inferior frontal gyrus, left orbital superior frontal gyrus, right caudate nucleus, right orbital superior frontal gyrus, right orbital middle frontal gyrus, left rectus gyrus, and left median cingulate and paracingulate gyrus of MDD patients. Compared with non-melancholic MDD, small-worldness was reduced and abnormal node attributes were in right orbital inferior frontal gyrus, left orbital superior frontal gyrus and right caudate nucleus of melancholic MDD. For correlation analysis, the 7th item score of the HRSD-17 (work and interest) was positively associated with increased node betweenness centrality (aBC) values in right orbital inferior frontal gyrus, while negatively associated with the decreased aBC in left orbital superior frontal gyrus. SVM analysis results showed that abnormal aBC in right orbital inferior frontal gyrus and left orbital superior frontal gyrus showed the highest accuracy of 81.0% (69/83), the sensitivity of 66.3%, and specificity of 85.2% for discriminating MDD patients with or without melancholic features.ConclusionThere is a significant difference in WM network changes between MDD patients with and without melancholic features.

Investigating Transfer Learning in Graph Neural Networks

Graph neural networks (GNNs) build on the success of deep learning models by extending them for use in graph spaces. Transfer learning has proven extremely successful for traditional deep learning problems, resulting in faster training and improved performance. Despite the increasing interest in GNNs and their use cases, there is little research on their transferability. This research demonstrates that transfer learning is effective with GNNs, and describes how source tasks and the choice of GNN impact the ability to learn generalisable knowledge. We perform experiments using real-world and synthetic data within the contexts of node classification and graph classification. To this end, we also provide a general methodology for transfer learning experimentation and present a novel algorithm for generating synthetic graph classification tasks. We compare the performance of GCN, GraphSAGE and GIN across both synthetic and real-world datasets. Our results demonstrate empirically that GNNs with inductive operations yield statistically significantly improved transfer. Further, we show that similarity in community structure between source and target tasks support statistically significant improvements in transfer over and above the use of only the node attributes.

A Novel Public Opinion Polarization Model Based on BA Network

At present, the polarization of online public opinion is becoming more frequent, and individuals actively participate in attitude interactions more and more frequently. Thus, online views have become the dominant force in current public opinion. However, the rapid fermentation of polarized public opinion makes it very easy for actual topic views to go to extremes. Significantly, negative information seriously affects the healthy development of the social opinion ecology. Therefore, it is beneficial to maintain national credibility, social peace, and stability by exploring the communication structure of online public opinions, analyzing the logical model of extreme public attitudes, and guiding the communication of public opinions in a timely and reasonable manner. Starting from the J–A model and BA network, this paper explores the specific attributes of individuals and opinion network nodes. By incorporating parameters such as individual conformity and the strength of individual online relationships, we established a model of online group attitude polarization, then conducted simulation experiments on the phenomenon of online opinion polarization. Through simulations, we found that individual conformity and the difference in environmental attitude greatly influence the direction of opinion polarization events. In addition, crowd mentality makes individuals spontaneously choose the side of a particular, extreme view, which makes it easier for polarization to form and reach its peak.

Network analysis of cattle movements in Chile: Implications for pathogen spread and control

Livestock movement between herds is one of the main routes for a pathogen to spread between herds. Understanding the contact structure, patterns and importance of specific individuals and classes can help assess dissemination risk, control, and surveillance measures. In this study, the cattle herd's contact structure in southern Chile was characterized based on the study of movement within one of the most important geographical regions for livestock production. Using Social Network Analysis, network measures of centrality, network cohesion and contact chains were estimated to identify essential features that can influence disease transmission among premises. Static and disaggregated by season networks were built for the entire set of nodes and for the Los Rios region animal flow only. For all networks, it was found that most nodes had a few movements; meanwhile, a small number of them were highly connected. Livestock markets have a powerful influence over the networks, being the most connected and central nodes, appearing as candidates for surveillance because of their possible role as super spreaders. Nodes attributes were associated with the likelihood of having high contact chains measures, such as herd class, location, and presence of markets anywhere on the contact chain. This work is the first deep analysis in livestock flows in Chile and shows the importance of animal movement data interpretation to assess pathogen risk spread in the country.

Monitoring and early warning model based on multi-dimensional structure of power grid data

With the continuous improvement of information technology, the monitoring of various conditions on power system is under continuous construction. The multiple monitoring system caused by network disaster is unable to achieve complete data exchange, so it is easy to lead to early warning of persistent conditions and chain conditions. In this paper, a DQN-based power monitoring node identification method for complex networks is proposed, and a key node identification algorithm for complex networks based on the Internet of Things is proposed. The reward function is improved and designed to integrate the influence of edge weights and node attributes. By acting on the power multi-dimensional structure monitoring and early warning of conditions, the key monitoring data are preferentially provided to other monitoring networks for analysis and prompt. Experiments show that in undirected ARPA networks, compared with degree centrality, betweenness centrality, Page Rank algorithm and two comprehensive methods, the recognition results are verified by using the evaluation method based on network robustness. The evaluation based on network robustness is improved by 12.5% compared with other algorithms; the node evaluation based on dynamics is more reasonable from the global attribute, which verifies the application effect of this algorithm in power multi-dimensional condition monitoring network.

Link-privacy preserving graph embedding data publication with adversarial learning

The inefficient utilization of ubiquitous graph data with combinatorial structures necessitates graph embedding methods, aiming at learning a continuous vector space for the graph, which is amenable to be adopted in traditional machine learning algorithms in favor of vector representations. Graph embedding methods build an important bridge between social network analysis and data analytics, as social networks naturally generate an unprecedented volume of graph data continuously. Publishing social network data not only brings benefit for public health, disaster response, commercial promotion, and many other applications, but also gives birth to threats that jeopardize each individual's privacy and security. Unfortunately, most existing works in publishing social graph embedding data only focus on preserving social graph structure with less attention paid to the privacy issues inherited from social networks. To be specific, attackers can infer the presence of a sensitive relationship between two individuals by training a predictive model with the exposed social network embedding. In this paper, we propose a novel link-privacy preserved graph embedding framework using adversarial learning, which can reduce adversary's prediction accuracy on sensitive links, while persevering sufficient non-sensitive information, such as graph topology and node attributes in graph embedding. Extensive experiments are conducted to evaluate the proposed framework using ground truth social network datasets.

Identifying influential spreaders in complex networks for disease spread and control

Identifying influential spreaders is an important task in controlling the spread of information and epidemic diseases in complex networks. Many recent studies have indicated that the identification of influential spreaders is dependent on the spreading dynamics. Finding a general optimal order of node importance ranking is difficult because of the complexity of network structures and the physical background of dynamics. In this paper, we use four metrics, namely, betweenness, degree, H-index, and coreness, to measure the central attributes of nodes for constructing the disease spreading models and target immunization strategies. Numerical simulations show that spreading processes based on betweenness centrality lead to the widest range of propagation and the smallest epidemic threshold for all six networks (including four real networks and two BA scale-free networks generated according to Barabasi–Albert algorithm). The target immunization strategy based on the betweenness centrality of nodes is the most effective for BA scale-free networks but displays poor immune effect for real networks in identifying the most important spreaders for disease control. The immunization strategy based on node degrees is the most effective for the four real networks. Findings show that the target immune strategy based on the betweenness centrality of nodes works best for standard scale-free networks, whereas that based on node degrees works best for other nonstandard scale-free networks. The results can provide insights into understanding the different metrics of measuring node importance in disease transmission and control.

Integration of Neighbor Topologies Based on Meta-Paths and Node Attributes for Predicting Drug-Related Diseases.

Identifying new disease indications for existing drugs can help facilitate drug development and reduce development cost. The previous drug–disease association prediction methods focused on data about drugs and diseases from multiple sources. However, they did not deeply integrate the neighbor topological information of drug and disease nodes from various meta-path perspectives. We propose a prediction method called NAPred to encode and integrate meta-path-level neighbor topologies, multiple kinds of drug attributes, and drug-related and disease-related similarities and associations. The multiple kinds of similarities between drugs reflect the degrees of similarity between two drugs from different perspectives. Therefore, we constructed three drug–disease heterogeneous networks according to these drug similarities, respectively. A learning framework based on fully connected neural networks and a convolutional neural network with an attention mechanism is proposed to learn information of the neighbor nodes of a pair of drug and disease nodes. The multiple neighbor sets composed of different kinds of nodes were formed respectively based on meta-paths with different semantics and different scales. We established the attention mechanisms at the neighbor-scale level and at the neighbor topology level to learn enhanced neighbor feature representations and enhanced neighbor topological representations. A convolutional-autoencoder-based module is proposed to encode the attributes of the drug–disease pair in three heterogeneous networks. Extensive experimental results indicated that NAPred outperformed several state-of-the-art methods for drug–disease association prediction, and the improved recall rates demonstrated that NAPred was able to retrieve more actual drug–disease associations from the top-ranked candidates. Case studies on five drugs further demonstrated the ability of NAPred to identify potential drug-related disease candidates.