Structural Deep Network Embedding

Networkembeddingisanimportantmethodtolearnlow-dimensional representations of vertexes in networks, aiming to capture and preserve the network structure. Almost all the existing network embeddingmethodsadoptshallowmodels. However,sincetheunderlyingnetworkstructureiscomplex, shallowmodelscannotcapture the highly non-linear network structure, resulting in sub-optimal network representations. Therefore, how to ﬁnd a method that is able to effectively capture the highly non-linear network structure and preserve the global and local structure is an open yet important problem. To solve this problem, in this paper we propose a StructuralDeepNetworkEmbeddingmethod,namelySDNE.More speciﬁcally, we ﬁrst propose a semi-supervised deep model, which has multiple layers of non-linear functions, thereby being able to capture the highly non-linear network structure. Then we propose to exploit the ﬁrst-order and second-order proximity jointly to preserve the network structure. The second-order proximity is used bytheunsupervisedcomponenttocapturetheglobalnetworkstructure. Whiletheﬁrst-orderproximityisusedasthesupervisedinformation in the supervised component to preserve the local network structure. By jointly optimizing them in the semi-supervised deep model, our method can preserve both the local and global network structureandisrobusttosparsenetworks. Empirically,weconduct the experiments on ﬁve real-world networks, including a language network, a citation network and three social networks. The results show that compared to the baselines, our method can reconstruct the original network signiﬁcantly better and achieves substantial gains in three applications, i.e. multi-label classiﬁcation, link prediction and visualization.

PDF

 ↳ Quickstart in
53
See all 6 implementations

Results from the Paper Add Remove

Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Graph Classification BP-fMRI-97 SDBN Accuracy 64.8% # 2
F1 63.7% # 3
Graph Classification HIV-DTI-77 SDBN Accuracy 65.9% # 2
F1 65.6% # 2
Graph Classification HIV-fMRI-77 SDBN Accuracy 66.5% # 3
F1 66.7% # 3