English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Betweenness centrality (BC) is one of the most used centrality measures fornetwork analysis, which seeks to describe the importance of nodes in a networkin terms of the fraction of shortest paths that pass through them. It is key tomany valuable applications, including community detection and networkdismantling. Computing BC scores on large networks is computationallychallenging due to high time complexity. Many approximation algorithms havebeen proposed to speed up the estimation of BC, which are mainlysampling-based. However, these methods are still prone to considerableexecution time on large-scale networks, and their results are often exacerbatedwhen small changes happen to the network structures. In this paper, we focus onidentifying nodes with high BC in a graph, since many application scenarios arebuilt upon retrieving nodes with top-k BC. Different from previous heuristicmethods, we turn this task into a learning problem and design anencoder-decoder based framework to resolve the problem. More specifcally, theencoder leverages the network structure to encode each node into an embeddingvector, which captures the important structural information of the node. Thedecoder transforms the embedding vector for each node into a scalar, whichcaptures the relative rank of this node in terms of BC. We use the pairwiseranking loss to train the model to identify the orders of nodes regarding theirBC. By training on small-scale networks, the learned model is capable ofassigning relative BC scores to nodes for any unseen networks, and thusidentifying the highly-ranked nodes. Comprehensive experiments on bothsynthetic and real-world networks demonstrate that, compared to representativebaselines, our model drastically speeds up the prediction without noticeablesacrifce in accuracy, and outperforms the state-of-the-art by accuracy onseveral large real-world networks.

Learning to Identify High Betweenness Centrality Nodes from Scratch: A Novel Graph Neural Network Approach