Deep Heterogeneous Social Network Alignment

Meng, Lin; Ren, Yuxiang; Zhang, Jiawei; Ye, Fanghua; Yu, Philip S.

doi:10.1109/cogmi48466.2019.00015

Cited by 3 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A few algorithms are used in multi-layer networks [Bindu et al, 2017, De Domenico et al, 2015, Tam et al, 2019, Jie et al, 2020 for abnormal nodes. The methods based on network alignment are used to multi-layer networks for anomaly subgraph detection [Sun et al, 2020, Liu et al, 2014, Heimann et al, 2018, Huynh et al, 2019, Yan et al, 2021, Sun et al, 2019, Wang et al, 2018b, Meng et al, 2019. Many federated learning methods on graphs are proposed to explore graph embedding vectors average [Xie et al, 2021, Li et al, 2020b, Zhang et al, 2021.…”

Section: Introductionmentioning

confidence: 99%

“…They want to complement the network structure and attribute information to achieve a better alignment effect. Moreover, there are many network alignment studies based on deep learning[Sun et al, 2019, Wang et al, 2018b, Meng et al, 2019. In addition, because the attribute information may be falsely fabricated or lost or hidden due to privacy, there are many alignment algorithms based only on structural information (BigAlign[Koutra et al, 2013], UMA, IONE[Liu et al, 2016], CrossMNA[Chu et al, 2019]).…”

mentioning

confidence: 99%

See 1 more Smart Citation

FadMan: Federated Anomaly Detection across Multiple Attributed Networks

Wu¹,

Zhang²,

Wang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Anomaly subgraph detection has been widely used in various applications, ranging from cyber attack in computer networks to malicious activities in social networks. Despite an increasing need for federated anomaly detection across multiple attributed networks, only a limited number of approaches are available for this problem. Federated anomaly detection faces two major challenges. One is that isolated data in most industries are restricted share with others for data privacy and security. The other is most of the centralized approaches training based on data integration. The main idea of federated anomaly detection is aligning private anomalies from local data owners on the public anomalies from the attributed network in the server through public anomalies to federate local anomalies. In each private attributed network, the detected anomaly subgraph is aligned with an anomaly subgraph in the public attributed network. The significant public anomaly subgraphs are selected for federated private anomalies while preventing local private data leakage. The proposed algorithm FadMan is a vertical federated learning framework for public node aligned with many private nodes of different features, and is validated on two tasks -correlated anomaly detection on multiple attributed networks and anomaly detection on an attributeless network -using five real-world datasets. In the first scenario, FadMan outperforms competitive methods by at least 12% accuracy at 10% noise level. In the second scenario, by analyzing the distribution of abnormal nodes, we find that the nodes of traffic anomalies are associated with the event of postgraduate entrance examination on the same day. 1 In this paper, we use 'graph ' and 'network', 'vertex' and 'node', and 'attribute' and 'feature' interchangeably. Preprint. Under review.

show abstract

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

FadMan: Federated Anomaly Detection across Multiple Attributed Networks

Wu¹,

Zhang²,

Wang³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Applications in many domains in the real world exhibit the favorable property of graph data structure, such as social networks [15], financial platforms [20] and bioinformatics [5]. Graph classification aims to identify the class labels of graphs in the dataset, which is an important problem for numerous applications.…”

Section: Introductionmentioning

confidence: 99%

Label Contrastive Coding based Graph Neural Network for Graph Classification

Ren

Bai

Zhang³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Graph classification is a critical research problem in many applications from different domains. In order to learn a graph classification model, the most widely used supervision component is an output layer together with classification loss (e.g.,cross-entropy loss together with softmax or margin loss). In fact, the discriminative information among instances are more fine-grained, which can benefit graph classification tasks. In this paper, we propose the novel Label Contrastive Coding based Graph Neural Network (LCGNN) to utilize label information more effectively and comprehensively. LCGNN still uses the classification loss to ensure the discriminability of classes. Meanwhile, LCGNN leverages the proposed Label Contrastive Loss derived from self-supervised learning to encourage instance-level intra-class compactness and interclass separability. To power the contrastive learning, LCGNN introduces a dynamic label memory bank and a momentum updated encoder. Our extensive evaluations with eight benchmark graph datasets demonstrate that LCGNN can outperform state-of-the-art graph classification models. Experimental results also verify that LCGNN can achieve competitive performance with less training data because LCGNN exploits label information comprehensively.

show abstract