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/

Global: Zendy Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Graph-based semi-supervised node classification (GraphSSC) has wideapplications, ranging from networking and security to data mining and machinelearning, etc. However, existing centralized GraphSSC methods are impracticalto solve many real-world graph-based problems, as collecting the entire graphand labeling a reasonable number of labels is time-consuming and costly, anddata privacy may be also violated. Federated learning (FL) is an emerginglearning paradigm that enables collaborative learning among multiple clients,which can mitigate the issue of label scarcity and protect data privacy aswell. Therefore, performing GraphSSC under the FL setting is a promisingsolution to solve real-world graph-based problems. However, existing FL methods1) perform poorly when data across clients are non-IID, 2) cannot handle datawith new label domains, and 3) cannot leverage unlabeled data, while all theseissues naturally happen in real-world graph-based problems. To address theabove issues, we propose the first FL framework, namely GraphFL, forsemi-supervised node classification on graphs. Our framework is motivated bymeta-learning methods. Specifically, we propose two GraphFL methods torespectively address the non-IID issue in graph data and handle the tasks withnew label domains. Furthermore, we design a self-training method to leverageunlabeled graph data. We adopt representative graph neural networks as GraphSSCmethods and evaluate GraphFL on multiple graph datasets. Experimental resultsdemonstrate that GraphFL significantly outperforms the compared FL baseline andGraphFL with self-training can obtain better performance.

GraphFL: A Federated Learning Framework for Semi-Supervised Node Classification on Graphs