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

Multi-view data containing complementary and consensus information canfacilitate representation learning by exploiting the intact integration ofmulti-view features. Because most objects in real world often have underlyingconnections, organizing multi-view data as heterogeneous graphs is beneficialto extracting latent information among different objects. Due to the powerfulcapability to gather information of neighborhood nodes, in this paper, we applyGraph Convolutional Network (GCN) to cope with heterogeneous-graph dataoriginating from multi-view data, which is still under-explored in the field ofGCN. In order to improve the quality of network topology and alleviate theinterference of noises yielded by graph fusion, some methods undertake sortingoperations before the graph convolution procedure. These GCN-based methodsgenerally sort and select the most confident neighborhood nodes for eachvertex, such as picking the top-k nodes according to pre-defined confidencevalues. Nonetheless, this is problematic due to the non-differentiable sortingoperators and inflexible graph embedding learning, which may result in blockedgradient computations and undesired performance. To cope with these issues, wepropose a joint framework dubbed Multi-view Graph Convolutional Network withDifferentiable Node Selection (MGCN-DNS), which is constituted of an adaptivegraph fusion layer, a graph learning module and a differentiable node selectionschema. MGCN-DNS accepts multi-channel graph-structural data as inputs and aimsto learn more robust graph fusion through a differentiable neural network. Theeffectiveness of the proposed method is verified by rigorous comparisons withconsiderable state-of-the-art approaches in terms of multi-view semi-supervisedclassification tasks.

Multi-view Graph Convolutional Networks with Differentiable Node Selection