BTWalk: Branching Tree Random Walk for Multi-Order Structured Network Embedding

Xiong, Hao; Yan, Junchi

doi:10.1109/tkde.2020.3029061

Cited by 12 publications

(14 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…iii) Multiplex network embedding: IONE [21] extends LINE from single networks to aligned pairwise networks, modeling both of in-degrees and out-degrees with concern for anchor links. BT-Walk [45] introduces binary-tree random walk for embedding and extends to cross-network by regularizers over anchor links. Both IONE and BTWalk can only deal with multiplex networks of two layers.…”

Section: Methodsmentioning

confidence: 99%

“…For projection heads h (д) v , we investigate three mostly used ones for contrastive learning on graphs, including identical mapping, linear mapping [8,48], and non-linear mapping [17,47]. Among them, identical mapping is the most commonly used one in NE for structure preserving [21,30,37,45]. Linear or non-linear mapping are believed to preserve more semantic information, e.g.…”

Section: 32mentioning

confidence: 99%

“…Network alignment [6,53] aims to predict anchor links between layers of multiplex networks. As a common protocol in line with [8,21,45], we randomly choose a specific portion of anchor links as the known seed matchings across layers, and predict the rest of anchor links according to the learned embeddings {X (д) } N д=1 .…”

Section: Network Alignmentmentioning

confidence: 99%

“…Between two layers, aligning neighborhoods of the nodes that behave consistently may help predict new inter-layer anchor links [3] and propagate true node labels across layers, while aligning the neighborhoods that disagree with each other may have just the opposite effects. However, most of the previous work [8,21,45] treat the inter-layer anchor links equally, which might not be reasonable enough. In our framework, we use an attention tensor to measure the agreement value of anchor links.…”

Section: Introductionmentioning

confidence: 98%

“…The problem of intra-layer learning for node embedding has been well studied by lots of existing works [16,30]. Among them, an augmented structural view is usually carefully designed to capture structure information of various granularity [14,30,45]. Recently, inspired by the progress of contrastive learning in visual representation learning that makes use of multiple views of a single picture (e.g.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Contrastive Multi-View Multiplex Network Embedding with Applications to Robust Network Alignment

Xiong

Yan

Pan

2021

Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

Despite its success in learning network node representations, network embedding is still relatively new for multiplex networks (MNs) with multiple types of edges. In such networks, the inter-layer anchor links are usually missing, which represent the alignment relations between nodes on different layers and are a crucial prerequisite for many cross-network applications like network alignment. For mining such anchor links between layers for MNs, multiplex network embedding (MNE) has become one of the most promising techniques. In this paper, we consider two problems for MNs: 1) edges can be missing to different extent, and data augmentation may mitigate this issue; 2) the known alignment anchor links between layers can be misleading since the behaviors of nodes on different layers are not always consistent, so the most informative ones should be emphasized compared with those misleading ones. However, most existing works neglect the two problems and simply 1) adopt one structural view for all the layers (e.g. random walk with the same window size) and 2) equally extract information from all the anchor links. We propose an end-to-end contrastive framework called cM 2 NE for MNE, utilizing multiple structural views for each layer and learning with several plug-in components for different scenarios. Through end-to-end optimization on three levels, the intra-view, inter-view, and inter-layer level, our framework achieves to select the fitted views for different layers and maximize the inter-layer mutual information by emphasizing those most informative anchor links. Extensive experimental results on real-world datasets for node classification and multi-network alignment show that our approach consistently outperforms peer methods. CCS CONCEPTS• Information systems → Data mining.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: 32mentioning

confidence: 99%

Section: Network Alignmentmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Contrastive Multi-View Multiplex Network Embedding with Applications to Robust Network Alignment

Xiong

Yan

Pan

2021

Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery &Amp; Data Mining

Self Cite

View full text Add to dashboard Cite

show abstract

How are You Affected? A Structural Graph Neural Network Model Predicting Individual Social Influence Status

Pan

2021

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

A subspace constraint based approach for fast hierarchical graph embedding

Chen

et al. 2023

World Wide Web

View full text Add to dashboard Cite

Hierarchy network, as a type of complex graphs, is widely used in many application scenarios such as social network analysis in web, human resource analysis in e-government, and product recommendation in ecommerce. Hierarchy preserving network embedding is a representation learning method that project nodes into feature space by preserving the hierarchy property of networks. Recently, researches on network embedding are devoted to mining hierarchical structures and profit a lot form it. Among these works, SpaceNE stands out of preserving hierarchy with the help of subspace constraint on the hierarchical subspace system. However, like all other existing works, SpaceNE is based on transductive learning method and is hard to generalize to new nodes. Besides, they have high time complexity and hard to be scalable to large-scale networks. This paper proposes an inductive method, FastHGE to learn node representation more efficiently and generalize to new nodes more easily. As SpaceNE, a hierarchy network is embedded into a hierarchical subspace tree. For upper communities, we exploit transductive learning by preserving inner-subspace proximity of subspace from the same ancestor. For extending to new nodes, we adopt inductive learning to learn representations of leaf nodes. The overall representation of a node is retrieved A Subspace Constraint based Approach for Fast Hierarchical Graph Embedding by concatenating the embedding vectors of all its ancestor communities and the leaf node. By learning the basis vectors of subspace, the computing cost is alleviated from updating many parameters of projection matrices as in SpaceNE. The performance evaluation experiments show that FastHGE outperforms with much fast speed and the same accuracy. For example, in the node classification, FastHGE is nearly 30 times faster than SpaceNE. The source code of FastHGE is available online.

show abstract

BTWalk: Branching Tree Random Walk for Multi-Order Structured Network Embedding

Cited by 12 publications

References 36 publications

Contrastive Multi-View Multiplex Network Embedding with Applications to Robust Network Alignment

Contrastive Multi-View Multiplex Network Embedding with Applications to Robust Network Alignment

How are You Affected? A Structural Graph Neural Network Model Predicting Individual Social Influence Status

A subspace constraint based approach for fast hierarchical graph embedding

Contact Info

Product

Resources

About