Name
Abstract | ||
|---|---|---|
Attributed network embedding (ANE) maps nodes in network into the low-dimensional space while preserving proximities of both node attributes and network topology. Existing methods for ANE integrated node attributes and network topology by three fusion strategies: the early fusion (EF), the synchronous fusion (SF) and the late fusion (LF). In fact, different fusion strategies have their own advantages and disadvantages. In this paper, we develop a dual fusion model named as DFANE. DFANE integrated the EF and the LF into a united framework, where the EF captures the latent complementarity and the LF extracts the distinctive information from node attributes and network topology. Extensive experiments on eight real-world networks have demonstrated the effectiveness and rationality of the DFANE. |
Year | DOI | Venue |
|---|---|---|
2020 | 10.1007/978-3-030-55130-8_8 | KSEM (1) |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Kunjie Dong | 1 | 0 | 0.34 |
| Lihua Zhou | 2 | 18 | 7.71 |
| Bing Kong | 3 | 16 | 1.67 |
| Junhua Zhou | 4 | 0 | 0.34 |