Name
Abstract | ||
|---|---|---|
Network embedding is an effective method to learn low-dimensional node vector representations which can well preserve the original network structures. Existing network embedding algorithms are mostly developed for a single network, which fail to learn generalized and comparable feature representations across networks. In this paper, we address a cross-network node classification problem by leveraging the abundant labeled information from a source network to help classify unlabeled nodes in a target network. A cross-network deep network embedding (CDNE) model is proposed to embed the nodes from the source network and the target network into a unified low-dimensional latent space. This model integrates deep network embedding and domain adaptation to learn label-discriminative and network-invariant node vector representations. The network structures, node attributes and node labels are leveraged collectively to learn similar hidden vector representations for similar nodes within a network and across different networks. Extensive experimental results demonstrate that the proposed model significantly outperforms the state-of-the-art related algorithms for node classification in the target network. |
Year | Venue | DocType |
|---|---|---|
2019 | arXiv: Social and Information Networks | Journal |
Volume | Citations | PageRank |
abs/1901.07264 | 0 | 0.34 |
References | Authors | |
0 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Xiao Shen | 1 | 0 | 0.34 |
| Fu-lai Chung | 2 | 244 | 34.50 |