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 Neural Networks (GNNs) require a large number of labeled graph samplesto obtain good performance on the graph classification task. The performance ofGNNs degrades significantly as the number of labeled graph samples decreases.To reduce the annotation cost, it is therefore important to develop graphaugmentation methods that can generate new graph instances to increase the sizeand diversity of the limited set of available labeled graph samples. In thiswork, we propose a novel mixup-based graph augmentation method, Graph DualMixup (GDM), that leverages both functional and structural information of thegraph instances to generate new labeled graph samples. GDM employs a graphstructural auto-encoder to learn structural embeddings of the graph samples,and then applies mixup to the structural information of the graphs in thelearned structural embedding space and generates new graph structures from themixup structural embeddings. As for the functional information, GDM appliesmixup directly to the input node features of the graph samples to generatefunctional node feature information for new mixup graph instances. Jointly, thegenerated input node features and graph structures yield new graph sampleswhich can supplement the set of original labeled graphs. Furthermore, wepropose two novel Balanced Graph Sampling methods to enhance the balanceddifficulty and diversity for the generated graph samples. Experimental resultson the benchmark datasets demonstrate that our proposed method substantiallyoutperforms the state-of-the-art graph augmentation methods when the labeledgraphs are scarce.

GDM: Dual Mixup for Graph Classification with Limited Supervision