Effect of homophily on network formation

Kim, Kibae; Altmann, Jörn

doi:10.1016/j.cnsns.2016.08.011

Cited by 46 publications

(31 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We incorporate homophily in this way because its influence on an incoming node's linking choices remains constant. Some authors incorporate homophily as a weighting of the degree of a node, and find, similar to our results here, that homophily "makes the rich even richer"[Kim and Altmann, 2017] and that homophily can lead to minority group members occupying less important places in the network[Karimi et al, 2018].8 One might also think that winning a credit attribution contest would make an existing node more "prestigious" and thus would increase its value in the eyes of incoming nodes. This is not included in the model, but would likely intensify the effects reported here.9 This HEG advantage exists in a model where information spreads by diffusion as well, in a simple contagion model where each discoverer tells their neighbors, who tell their neighbors that have not already heard of the discovery, and so on (with people receiving conflicting information choosing to assign credit randomly).…”

supporting

confidence: 82%

Priority and privilege in scientific discovery

Rubin

Schneider

2021

Studies in History and Philosophy of Science Part A

View full text Add to dashboard Cite

The priority rule in science has been interpreted as a behavior regulator for the scientific community, which benefits society by adequately structuring the distribution of intellectual labor across pre-existing research programs. Further, it has been lauded as part of society's "grand reward scheme" because it fairly rewards people for the benefits they produce. But considerations about how news of scientific developments spreads throughout a scientific community at large suggest that the priority rule is something else entirely, which can disadvantage historically underrepresented or otherwise marginalized social groups.

show abstract

supporting

confidence: 82%

Priority and privilege in scientific discovery

Rubin

Schneider

2021

Studies in History and Philosophy of Science Part A

View full text Add to dashboard Cite

show abstract

“…From characteristics of real-world datasets [22] and empirical results of the state-of-the-art [17,24,49], we argue that a high-quality network alignment solution should respect: (R1) Consistency: Structural consistency and attribute consistency shall be respected since these constraints help to find true anchor links [17,49]. False positives could hamper downstream applications, such as personalized advertisement and friend suggestion.…”

Section: A Motivationmentioning

confidence: 99%

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)

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

“…Addressing the problem formulated above is indeed a challenging task. The alignment model should consider the structure and attribute consistencies in an efficient manner, as these are essential guidelines for a high-quality alignment solution [35]. The alignment model should also be aware of structure and feature consistency violations (noise).…”

Section: Framework Overviewmentioning

confidence: 99%

Network Alignment with Holistic Embeddings

Huynh

Duong

Nguyen

et al. 2021

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

Network alignment is the task of identifying topologically and semantically similar nodes across (two) different networks. It plays an important role in various applications ranging from social network analysis to bioinformatic network interactions. However, existing alignment models either cannot handle large-scale graphs or fail to leverage different types of network information or modalities. In this paper, we propose a novel end-to-end alignment framework that can leverage different modalities to compare and align network nodes in an efficient way. In order to exploit the richness of the network context, our model constructs multiple embeddings for each node, each of which captures one modality or type of network information. We then design a late-fusion mechanism to combine the learned embeddings based on the importance of the underlying information. Our fusion mechanism allows our model to be adapted to various types of structure of the input network. Experimental results show that our technique outperforms state-of-the-art approaches in terms of accuracy on real and synthetic datasets, while being robust against various noise factors.

show abstract

Effect of homophily on network formation

Cited by 46 publications

References 33 publications

Priority and privilege in scientific discovery

Priority and privilege in scientific discovery

Adaptive Network Alignment with Unsupervised and Multi-order Convolutional Networks

Network Alignment with Holistic Embeddings

Contact Info

Product

Resources

About