A comparative study on network alignment techniques

Huynh, Thanh Trung; Toan, Nguyen Thanh; Vinh, Tong Van; Hoang, Thanh Dat; Thang, Duong Chi; Hùng, Nguyễn Quốc; Sattar, Abdul

doi:10.1016/j.eswa.2019.112883

Cited by 51 publications

(23 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This above formulation has two advantages. (i) it enables scalable calculation, since anchor links can be inferred from the alignment matrix through ranking rules [17,23]; and (ii) it provides flexibility since node pairings can be one-to-many, which is important for differently sized networks [21,29,38].…”

Section: Problem Definition (Alignment Matrix Computation)mentioning

confidence: 99%

“…Another problem is network reconciliation, where two different networks are connected for data integration [25,30]. Our work solves the recent network alignment problem, which finds anchor links between an original network and its variants [2,33,36,38].…”

Section: Related Workmentioning

confidence: 99%

“…Spectral (Matrix factorisation) approach. Many approaches use matrix factorisation to compute the alignment matrix directly [38]. The classic IsoRank [36] propagates the pairwise node similarity along with structural consistency over the network.…”

Section: A Network Alignmentmentioning

confidence: 99%

“…Network alignment, the task of pairing nodes between two isomorphic or nearisomorphic networks such that the paired nodes are similar w.r.t. network structure and attribute information, is the foundation to extract valuable knowledge in these applications [38]. For example, in social networks, the detection of different accounts (e.g., Facebook, Twitter) of the same user facilitates personalized advertisement, friend suggestion, and content recommendation [39,40,45,46,49].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Adaptive Network Alignment with Unsupervised and Multi-order Convolutional Networks

Huynh

Vinh

Tâm

et al. 2020

2020 IEEE 36th International Conference on Data Engineering (ICDE)

Self Cite

View full text Add to dashboard Cite

Network alignment is the problem of pairing nodes between two graphs such that the paired nodes are structurally and semantically similar. A well-known application of network alignment is to identify which accounts in different social networks belong to the same person. Existing alignment techniques, however, lack scalability, cannot incorporate multi-dimensional information without training data, and are limited in the consistency constraints enforced by an alignment. In this paper, we propose a fully unsupervised network alignment framework based on a multi-order embedding model. The model learns the embeddings of each node using a graph convolutional neural representation, which we prove to satisfy consistency constraints. We further design a data augmentation method and a refinement mechanism to make the model adaptive to consistency violations and noise. Extensive experiments on real and synthetic datasets show that our model outperforms state-of-the-art alignment techniques. We also demonstrate the robustness of our model against adversarial conditions, such as structural noises, attribute noises, graph size imbalance, and hyper-parameter sensitivity.

show abstract

Section: Problem Definition (Alignment Matrix Computation)mentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: A Network Alignmentmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Adaptive Network Alignment with Unsupervised and Multi-order Convolutional Networks

Huynh

Vinh

Tâm

et al. 2020

2020 IEEE 36th International Conference on Data Engineering (ICDE)

Self Cite

View full text Add to dashboard Cite

show abstract

“…These include matrix factorization-based methods as well as more recent neural network-based methods [ 31 , 32 ]. Embedding methods provide a vectorized representation for each gene/protein and are often faster than other options, which can be critical when dealing with analysing networks [ 33 ]. Additionally, the learned embeddings are often applicable for downstream analysis as the method provides a numeric representation of the genes that can be fed into a machine learning algorithm, for example, while capturing information about how it is positioned in a network.…”

Section: Introductionmentioning

confidence: 99%

Juxtapose: a gene-embedding approach for comparing co-expression networks

et al. 2021

View full text Add to dashboard Cite

Background Gene co-expression networks (GCNs) are not easily comparable due to their complex structure. In this paper, we propose a tool, Juxtapose, together with similarity measures that can be utilized for comparative transcriptomics between a set of organisms. While we focus on its application to comparing co-expression networks across species in evolutionary studies, Juxtapose is also generalizable to co-expression network comparisons across tissues or conditions within the same species. Methods A word embedding strategy commonly used in natural language processing was utilized in order to generate gene embeddings based on walks made throughout the GCNs. Juxtapose was evaluated based on its ability to embed the nodes of synthetic structures in the networks consistently while also generating biologically informative results. Evaluation of the techniques proposed in this research utilized RNA-seq datasets from GTEx, a multi-species experiment of prefrontal cortex samples from the Gene Expression Omnibus, as well as synthesized datasets. Biological evaluation was performed using gene set enrichment analysis and known gene relationships in literature. Results We show that Juxtapose is capable of globally aligning synthesized networks as well as identifying areas that are conserved in real gene co-expression networks without reliance on external biological information. Furthermore, output from a matching algorithm that uses cosine distance between GCN embeddings is shown to be an informative measure of similarity that reflects the amount of topological similarity between networks. Conclusions Juxtapose can be used to align GCNs without relying on known biological similarities and enables post-hoc analyses using biological parameters, such as orthology of genes, or conserved or variable pathways. Availability A development version of the software used in this paper is available at https://github.com/klovens/juxtapose

show abstract

Embedding-Based Network Alignment Using Neural Tensor Networks

Ning

et al. 2021

Knowledge Science, Engineering and Management

View full text Add to dashboard Cite

A comparative study on network alignment techniques

Cited by 51 publications

References 24 publications

Adaptive Network Alignment with Unsupervised and Multi-order Convolutional Networks

Adaptive Network Alignment with Unsupervised and Multi-order Convolutional Networks

Juxtapose: a gene-embedding approach for comparing co-expression networks

Embedding-Based Network Alignment Using Neural Tensor Networks

Contact Info

Product

Resources

About