Neighbor Nodes Research Articles

DGA-GNN: Dynamic Grouping Aggregation GNN for Fraud Detection

Fraud detection has increasingly become a prominent research field due to the dramatically increased incidents of fraud. The complex connections involving thousands, or even millions of nodes, present challenges for fraud detection tasks. Many researchers have developed various graph-based methods to detect fraud from these intricate graphs. However, those methods neglect two distinct characteristics of the fraud graph: the non-additivity of certain attributes and the distinguishability of grouped messages from neighbor nodes. This paper introduces the Dynamic Grouping Aggregation Graph Neural Network (DGA-GNN) for fraud detection, which addresses these two characteristics by dynamically grouping attribute value ranges and neighbor nodes. In DGA-GNN, we initially propose the decision tree binning encoding to transform non-additive node attributes into bin vectors. This approach aligns well with the GNN’s aggregation operation and avoids nonsensical feature generation. Furthermore, we devise a feedback dynamic grouping strategy to classify graph nodes into two distinct groups and then employ a hierarchical aggregation. This method extracts more discriminative features for fraud detection tasks. Extensive experiments on five datasets suggest that our proposed method achieves a 3% ~ 16% improvement over existing SOTA methods. Code is available at https://github.com/AtwoodDuan/DGA-GNN.

Distributed Manifold Hashing for Image Set Classification and Retrieval

Conventional image set methods typically learn from image sets stored in one location. However, in real-world applications, image sets are often distributed or collected across different positions. Learning from such distributed image sets presents a challenge that has not been studied thus far. Moreover, efficiency is seldom addressed in large-scale image set applications. To fulfill these gaps, this paper proposes Distributed Manifold Hashing (DMH), which models distributed image sets as a connected graph. DMH employs Riemannian manifold to effectively represent each image set and further suggests learning hash code for each image set to achieve efficient computation and storage. DMH is formally formulated as a distributed learning problem with local consistency constraint on global variables among neighbor nodes, and can be optimized in parallel. Extensive experiments on three benchmark datasets demonstrate that DMH achieves highly competitive accuracies in a distributed setting and provides faster classification and retrieval than state-of-the-arts.

Structural Information Enhanced Graph Representation for Link Prediction

Link prediction is a fundamental task of graph machine learning, and Graph Neural Network (GNN) based methods have become the mainstream approach due to their good performance. However, the typical practice learns node representations through neighborhood aggregation, lacking awareness of the structural relationships between target nodes. Recently, some methods have attempted to address this issue by node labeling tricks. However, they still rely on the node-centric neighborhood message passing of GNNs, which we believe involves two limitations in terms of information perception and transmission for link prediction. First, it cannot perceive long-range structural information due to the restricted receptive fields. Second, there may be information loss of node-centric model on link-centric task. In addition, we empirically find that the neighbor node features could introduce noise for link prediction. To address these issues, we propose a structural information enhanced link prediction framework, which involves removing the neighbor node features while fitting neighborhood graph structures more focused through GNN. Furthermore, we introduce Binary Structural Transformer (BST) to encode the structural relationships between target nodes, complementing the deficiency of GNN. Our approach achieves remarkable results on multiple popular benchmarks, including ranking first on ogbl-ppa, ogbl-citation2 and Pubmed.

SGFormer: Semantic Graph Transformer for Point Cloud-Based 3D Scene Graph Generation

In this paper, we propose a novel model called SGFormer, Semantic Graph TransFormer for point cloud-based 3D scene graph generation. The task aims to parse a point cloud-based scene into a semantic structural graph, with the core challenge of modeling the complex global structure. Existing methods based on graph convolutional networks (GCNs) suffer from the over-smoothing dilemma and can only propagate information from limited neighboring nodes. In contrast, SGFormer uses Transformer layers as the base building block to allow global information passing, with two types of newly-designed layers tailored for the 3D scene graph generation task. Specifically, we introduce the graph embedding layer to best utilize the global information in graph edges while maintaining comparable computation costs. Furthermore, we propose the semantic injection layer to leverage linguistic knowledge from large-scale language model (i.e., ChatGPT), to enhance objects' visual features. We benchmark our SGFormer on the established 3DSSG dataset and achieve a 40.94% absolute improvement in relationship prediction's R@50 and an 88.36% boost on the subset with complex scenes over the state-of-the-art. Our analyses further show SGFormer's superiority in the long-tail and zero-shot scenarios. Our source code is available at https://github.com/Andy20178/SGFormer.

DiG-In-GNN: Discriminative Feature Guided GNN-Based Fraud Detector against Inconsistencies in Multi-Relation Fraud Graph

Fraud detection on multi-relation graphs aims to identify fraudsters in graphs. Graph Neural Network (GNN) models leverage graph structures to pass messages from neighbors to the target nodes, thereby enriching the representations of those target nodes. However, feature and structural inconsistency in the graph, owing to fraudsters' camouflage behaviors, diminish the suspiciousness of fraud nodes which hinders the effectiveness of GNN-based models. In this work, we propose DiG-In-GNN, Discriminative Feature Guided GNN against Inconsistency, to dig into graphs for fraudsters. Specifically, we use multi-scale contrastive learning from the perspective of the neighborhood subgraph where the target node is located to generate guidance nodes to cope with the feature inconsistency. Then, guided by the guidance nodes, we conduct fine-grained neighbor selection through reinforcement learning for each neighbor node to precisely filter nodes that can enhance the message passing and therefore alleviate structural inconsistency. Finally, the two modules are integrated together to obtain discriminable representations of the nodes. Experiments on three fraud detection datasets demonstrate the superiority of the proposed method DiG-In-GNN, which obtains up to 20.73% improvement over previous state-of-the-art methods. Our code can be found at https://github.com/GraphBerry/DiG-In-GNN.

A novel distributed bearing‐only target tracking algorithm for underwater sensor networks with resource constraints

AbstractUnderwater sensor networks hold immense potential for advancing the field of underwater target tracking, yet they encounter significant resource constraints stemming from energy storage and communication methods. In order to balance tracking accuracy and energy consumption, the authors present a distributed bearing‐only target tracking algorithm that can be used in underwater sensor networks with resource constraints. Anchored in the diffusion cubature information filter framework, this algorithm achieves fusion for non‐linear bearing measurements and state estimation. During the incremental update stage, individual nodes leverage the Posterior Cramer‐Rao Lower Bound as a metric for tracking performance. Subsequently, a strategy for selecting neighbouring nodes is introduced, ensuring tracking accuracy while efficiently kerbing energy consumption. In the diffusion update stage, a multi‐threshold event triggering mechanism is employed to partially diffuse the intermediate estimation. Additionally, an adaptive convex combination weight is proposed for cases involving partial diffusion. Through theoretical analysis, the asymptotic unbiasedness and convergence of the algorithm have been proven. Through Monte Carlo simulation experiments, the authors verify that the algorithm is superior to existing algorithms. Furthermore, the algorithm significantly reduces energy consumption in information interaction, minimising tracking accuracy loss.

PCSSHO: An implementation of Percentage of Circle Search and Spotted Hyena Optimizer for multi‐constraint load balancing and routing framework in FANET

SummaryFlying ad hoc network (FANET) is a technology that has seen tremendous growth in the past years due to its application in various military and civil developments. Conventional topology‐aided routing schemes are not suitable for large‐scale FANETs. This is because of the higher mobility rate in UAVs with the architecture differences. Hence, several path entries in routing become invalid and the neighboring nodes can be occupied before the interruption. Hence, it is very important to solve the afore‐explained issues in FANET. Initially, the task is assigned to the UAVs. These tasks are performed by satisfying the demands such as timing, congestion, and energy to attain the effective load‐balancing performance. Then, the optimal clustering and cluster head (CH) selection are performed using the developed Percentage of Circle Search and Spotted Hyena Optimizer (PCSSHO) in the communication path. Here, the feature selection is done based on constraints like position, speed, moving direction, height variation, link quality, and inter‐ and intra‐cluster distance. Then, the developed PCSSHO model performs routing by considering different constraints such as “end‐to‐end delay, delivery ratio, power consumption, and link quality.” Thus, the recommenced load balancing model in FANET secures an enhanced performance rate than the conventional load balancing frameworks.

ASG-HOMGAT: a high-order multi-head graph attention network with adaptive small graph structure for rolling bearing fault diagnosis

Traditional Euclidean spatial data processing is difficult to capture the inherent relationships of unstructured data such as bearing vibration signals. Representing vibration signals in graphical form helps to preserve their topological structure and temporal information. Secondly, most existing graph convolutional network methods are based on large graph structured data, which incurs certain memory overhead when aggregating high-order neighborhood node information and ignores important information between samples in the global graph structure. To address these issues, this paper proposes a high-order multi-head graph attention network based on an adaptive small graph structure (ASG-HOMGAT) for fault diagnosis of rolling bearings. Firstly, the adaptive preprocessing layer is used to adaptively denoise and compress the one-dimensional time-domain vibration signal, generating small rule graph data with topological structure. Then, these small graph structured data samples are input into a higher-order graph neural network, which aggregates features from multiple higher-order neighborhoods to achieve richer feature representations and fully explore the intrinsic correlation between samples. Finally, these features are aggregated into a reinforced representation of graph nodes through a multi head attention mechanism, and a SoftMax classifier is used for fault classification. The experimental results show that the ASG-HOMGAT method has better performance compared to mainstream graph neural network diagnostic models. The code and model will be released at: https://github.com/ding-ss/ASG-HOMGAT.

Simple and deep graph attention networks

Graph Attention Networks (GATs) and Graph Convolutional Neural Networks (GCNs) are two state-of-the-art architectures in Graph Neural Networks (GNNs). It is well known that both models suffer from performance degradation when more GNN layers are stacked, and many works have been devoted to address this problem. We notice that main research efforts in the line focus on the GCN models, and their techniques cannot well fit the GAT models due to the inherent difference between these two architectures. In GAT, the attention mechanism is limited as it ignores the overwhelming propagation from certain nodes as the number of layers increases. To sufficiently utilize the expressive power of GAT, we propose a new version of GAT named Layer-wise Self-adaptive GAT (LSGAT), which can effectively alleviate the oversmoothing issue in deep GAT and is strictly more expressive than GAT. We redesign the attention coefficients computation mechanism adaptively adjusted by layer depth, which considers both immediate neighboring and non-adjacent nodes from a global view. The experimental evaluation confirms that LSGAT consistently achieves better results on node classification tasks over relevant counterparts.

MINDG: a drug-target interaction prediction method based on an integrated learning algorithm.

Drug-target interaction (DTI) prediction refers to the prediction of whether a given drug molecule will bind to a specific target and thus exert a targeted therapeutic effect. Although intelligent computational approaches for drug target prediction have received much attention and made many advances, they are still a challenging task that requires further research. The main challenges are manifested as follows: (i) most graph neural network-based methods only consider the information of the first-order neighboring nodes (drug and target) in the graph, without learning deeper and richer structural features from the higher-order neighboring nodes. (ii) Existing methods do not consider both the sequence and structural features of drugs and targets, and each method is independent of each other, and cannot combine the advantages of sequence and structural features to improve the interactive learning effect. To address the above challenges, a Multi-view Integrated learning Network that integrates Deep learning and Graph Learning (MINDG) is proposed in this study, which consists of the following parts: (i) a mixed deep network is used to extract sequence features of drugs and targets, (ii) a higher-order graph attention convolutional network is proposed to better extract and capture structural features, and (iii) a multi-view adaptive integrated decision module is used to improve and complement the initial prediction results of the above two networks to enhance the prediction performance. We evaluate MINDG on two dataset and show it improved DTI prediction performance compared to state-of-the-art baselines. https://github.com/jnuaipr/MINDG.

Dense Multi-Scale Graph Convolutional Network for Knee Joint Cartilage Segmentation.

In this paper, we propose a dense multi-scale adaptive graph convolutional network (DMA-GCN) method for automatic segmentation of the knee joint cartilage from MR images. Under the multi-atlas setting, the suggested approach exhibits several novelties, as described in the following. First, our models integrate both local-level and global-level learning simultaneously. The local learning task aggregates spatial contextual information from aligned spatial neighborhoods of nodes, at multiple scales, while global learning explores pairwise affinities between nodes, located globally at different positions in the image. We propose two different structures of building models, whereby the local and global convolutional units are combined by following an alternating or a sequential manner. Secondly, based on the previous models, we develop the DMA-GCN network, by utilizing a densely connected architecture with residual skip connections. This is a deeper GCN structure, expanded over different block layers, thus being capable of providing more expressive node feature representations. Third, all units pertaining to the overall network are equipped with their individual adaptive graph learning mechanism, which allows the graph structures to be automatically learned during training. The proposed cartilage segmentation method is evaluated on the entire publicly available Osteoarthritis Initiative (OAI) cohort. To this end, we have devised a thorough experimental setup, with the goal of investigating the effect of several factors of our approach on the classification rates. Furthermore, we present exhaustive comparative results, considering traditional existing methods, six deep learning segmentation methods, and seven graph-based convolution methods, including the currently most representative models from this field. The obtained results demonstrate that the DMA-GCN outperforms all competing methods across all evaluation measures, providing DSC=95.71% and DSC=94.02% for the segmentation of femoral and tibial cartilage, respectively.

General protocol for predicting outbreaks of infectious diseases in social networks

Epidemic spreading on social networks with quenched connections is strongly influenced by dynamic correlations between connected nodes, posing theoretical challenges in predicting outbreaks of infectious diseases. The quenched connections introduce dynamic correlations, indicating that the infection of one node increases the likelihood of infection among its neighboring nodes. These dynamic correlations pose significant difficulties in developing comprehensive theories for threshold determination. Determining the precise epidemic threshold is pivotal for diseases control. In this study, we propose a general protocol for accurately determining epidemic thresholds by introducing a new set of fundamental conditions, where the number of connections between individuals of each type remains constant in the stationary state, and by devising a rescaling method for infection rates. Our general protocol is applicable to diverse epidemic models, regardless of the number of stages and transmission modes. To validate our protocol’s effectiveness, we apply it to two widely recognized standard models, the susceptible–infected–recovered-susceptible model and the contact process model, both of which have eluded precise threshold determination using existing sophisticated theories. Our results offer essential tools to enhance disease control strategies and preparedness in an ever-evolving landscape of infectious diseases.

An Optimized Link State Routing Protocol with a Blockchain Framework for Efficient Video-Packet Transmission and Security over Mobile Ad-Hoc Networks

A mobile ad-hoc network (MANET) necessitates appropriate routing techniques to enable optimal data transfer. The selection of appropriate routing protocols while utilizing the default settings is required to solve the existing problems. To enable effective video streaming in MANETs, this study proposes a novel optimized link state routing (OLSR) protocol that incorporates a deep-learning model. Initially, the input videos are collected from the Kaggle dataset. Then, the black-hole node is detected using a novel twin-attention-based dense convolutional bidirectional gated network (SA_ DCBiGNet) model. Next, the neighboring nodes are analyzed using trust values, and routing is performed using the extended osprey-aided optimized link state routing protocol (EO_OLSRP) technique. Similarly, the extended osprey optimization algorithm (EOOA) selects the optimal feature based on parameters such as node stability and link stability. Finally, blockchain storage is included to improve the security of MANET data using interplanetary file system (IPFS) technology. Additionally, the proposed blockchain system is validated utilizing a consensus technique based on delegated proof-of-stake (DPoS). The proposed method utilizes Python and it is evaluated using data acquired from various mobile simulator models accompanied by the NS3 simulator. The proposed model performs better with a packet-delivery ratio (PDR) of 91.6%, average end delay (AED) of 23.6 s, and throughput of 2110 bytes when compared with the existing methods which have a PDR of 89.1%, AED of 22 s, and throughput of 1780 bytes, respectively.

QQAR: A Q-learning-based QoS-aware routing for IoMT-enabled wireless body area networks for smart healthcare

Recently, the Internet of Medical Things (IoMT) has gained broad acclaim in the academic and industrial sectors due to its use in integrating medical devices and healthcare systems. Wireless body area networks (WBANs) are a promising technology in smart healthcare for real-time patient monitoring and health data analysis by medical professionals. Efficient data transmission is essential in ensuring reliable and timely delivery of healthcare facilities. Healthcare services require reliable, delay-sensitive, energy- and congestion-aware communication, which makes it challenging to design a routing protocol. To address quality of services (QoS) in IoMT-enabled WBANs for healthcare, we propose Q-learning-based quality of services (QoS) aware routing (QQAR). Firstly, we used two-hop-based link reliability estimation to ensure link reliability from sources to sink nodes. Routing decisions using two-hop neighbor information guarantees successful packet delivery. Secondly, we differentiated the packets based on traffic priority, which assures the delivery of critical packets even in a congested network. Particularly, considering two-hop node velocity with energy balancing enhanced the guarantee of timely packet delivery before deadlines. In addition, the multi-sink approach enhanced network reliability. Finally, we utilized Q-learning to select suitable neighbor nodes for packet forwarding in a decentralized fashion. A multi-sink load balance and congestion control mechanism adjusts the routing decisions in the QQAR to avoid congestion. Our simulation and comparison results showed that QQAR outperforms the existing routing protocols across various performance metrics under distinct network conditions.

Trade strategy with the countries along the Ice Silk Road: a Rabin model-based perspective

ABSTRACT Since the Ice Silk Road (ISR) was put forward, its development prospect has become the focus of academic circles. To study the influence mechanism of ‘fairness preference’ on the trade cooperation of countries along the ISR, this paper constructs the trade cooperation network based on the bilateral trade data of countries along the ISR, establishes the theoretical model of network game based on Rabin model, and uses the simulation method to test the model. The results show that: (1) the sense of national cognitive fairness does have an impact on the trade cooperation of the ISR; (2) the ‘eye for an eye, tooth for a tooth’ strategy is not an advantage strategy in the trade cooperation of the ISR; (3) the most significant factor affecting the trade cooperation of countries along the ISR is the neighbor node in the trade network and the social public opinion pressure of cooperation in the network.

Developing a method for the detection and identification of rootstock blockchain network nodes

The object of this study is the protocol for detecting nodes in the Rootstock blockchain network and crawling tools. Node discovery protocols are the foundation of any decentralized peer-to-peer network. In blockchain systems, full nodes store and maintain a complete copy of all transactions performed in the network. However, they do not store information about all other nodes, such as their IDs or IP addresses. Each node usually maintains an incomplete list of nodes to which it connects to exchange blockchain data. In decentralized networks, nodes join and leave the network and their IP addresses can change, making it impractical to maintain a complete, up-to-date list of all nodes. Therefore, the only way to get a list of all nodes in the network is to poll each node sequentially. The developed method involves sending specially formed messages to nodes to obtain their neighbors. The graph search algorithm is used to traverse all received neighboring nodes. This allows one to consistently detect all network nodes. Identifying the desired sequence of messages requires a preliminary analysis of the node software RSKj in the part of node discovery protocol. Effectiveness of the proposed method was verified using the developed software and an experiment in the main network. 6 verification nodes were deployed in different physical locations and at different times. All test nodes were detected in less than 10 minutes. The developed method found 222 nodes that have 209 unique IP addresses. Results of this study show how to perform analysis of node discovery protocol. They provide the means to obtain information about the available nodes of the Rootstock blockchain system, enabling the analysis of both the blockchain network in general and individual node

Fuzzy logic‐based trusted routing protocol using vehicular cloud networks for smart cities

AbstractDue to the characteristics of vehicular ad hoc networks, the increased mobility of nodes and the inconsistency of wireless communication connections pose significant challenges for routing. As a result, researchers find it to be a fascinating topic to study. Furthermore, since these networks are vulnerable to various assaults, providing an authentication method between the source and destination nodes is crucial. How to route in such networks more efficiently, taking into account node mobility characteristics and accompanying massive historical data, is still a matter of discussion. Fuzzy logic‐based Trusted Routing Protocol for vehicular cloud networks (FTRP) is proposed in this study that determines the secure path for data dissemination. Fuzzy Logic determines the node candidacy value and selects or rejects a path accordingly. The cloud assigns a confidence score to each vehicle based on the data it collects from nodes after each interaction. Our study identifies the secure path on the basis of trust along with factors such as speed, closeness to other nodes, signal strength and distance from the neighbouring nodes. Simulations of the novel protocol demonstrate that it can keep the packet delivery ratio high with little overhead and low delay. FTRP has significant implications for deploying Vehicular Cloud Networks using electric vehicle technologies in smart cities. The routing data is collected with the help of Internet of Technology (IOT) sensors. The information is transmitted between vehicles using IOT gateways.

A Bayesian approach using spatiotemporal features for suitable next hop selection in opportunistic networks

SummaryOpportunistic network (OppNet) belongs to the category of Mobile Ad‐hoc Networks (MANETs), a kind of Delay Tolerant Network (DTN), where the wireless nodes are completely mobile and the data transmission routes are dynamic. The major challenge in developing a routing model for such a network is the unpredictable nature of the movement of the nodes. In this paper, a spatiotemporal prediction model based on human mobility pattern is proposed using Bayesian posterior probability (BPPR) where several clusters are identified within the network and the day and time duration of nodes visiting those clusters are recorded. The Bayesian posterior probability is then used to determine the probability of the neighbor node visiting the destination's cluster. If the calculated probability for that node is higher than a specified threshold, the packet will be forwarded. A comparison of the results obtained on simulation is made with benchmark models—Epidemic, Prophet, HBPR, EDR, NexT, and EBC, to name a few, where it is found that on average the proposed model outperforms the other models in terms of delivery probability, hop count and number of messages dropped by around 23.89%, 24.8%, 24.4%, 37%, 11%, and 42% respectively, with varying number of nodes, TTL, message generation interval, and buffer size. Similar improvements have been observed in terms of the other two metrics. In terms of overhead ratio, the proposed model outperforms Epidemic, Prophet, HBPR, NexT, and EBC. However, as the number of nodes and TTL are varied, BPPR performs better than NexT by around 9% and 12%, respectively, in terms of average latency.

CR-TransR: A Knowledge Graph Embedding Model for Cultural Domain

As a combination of information computing technology and the cultural field, cultural computing is gaining more attention. The knowledge graph is also gradually applied as a particular data structure in the cultural area. Based on the domain knowledge graph data of the Beijing Municipal Social Science Project “Mining and Utilization of Cultural Resources in the Ancient Capital of Beijing,” this article proposes a graph representation learning model CR-TransR that integrates cultural attributes. Through the analysis of the data in the cultural field of the ancient capital of Beijing, a cultural feature dictionary is constructed, and a domain-specific feature matrix is constructed in the form of word vector splicing. The feature matrix is used to constrain the embedding graph model TransR, and then the feature matrix and the TransR model are jointly trained to complete the embedded expression of the knowledge graph. Finally, a comparative experiment is carried out on the Beijing ancient capital cultural knowledge graph dataset and the effects of the classic graph embedding algorithms TransE, TransH, and TransR. At the same time, we try to reproduce the embedding method with the core idea of neighbor node information aggregation as the core idea, and CRTransR are compared. The experimental tasks include link prediction and triplet classification, and the experimental results show that the CRTransR model performs better.

Unifying heterogeneous and homogeneous relations for personalized compatibility modeling

In recent years, the fashion industry has developed rapidly, leading to an increasing demand for personalized clothing matching. Although noteworthy progress has been made by existing work with advanced graph neural networks, there still remain three prominent challenges: comprehensive entity relation modeling, adaptive neighbor node sampling, and thorough characteri-zation of fashion items based on multi-modalities. To solve the above problems, we propose a novel personalized compatibility modeling framework, which explores both heterogeneous (i.e., item-user and user-item) and homogeneous (i.e., item-item and user-user) relation graphs with multi-modalities. Moreover, we design a correlation-guided neighbor sampling strategy to learn the correlation-aware subgraphs, which plays a crucial role in facilitating the exploration of compatibility relationships or personalized preferences. Finally, we conduct extensive experiments on two publicly available fashion datasets to demonstrate the superiority of our proposed model over existing benchmark methods.