Name
Abstract | ||
|---|---|---|
This paper attempts to addresses the task of node classification in social networks. As we know, node classification in social networks is an important challenge for understanding the underlying graph with the linkage structure and node features. Compared with the traditional classification problem, we should not only use the node features, but also consider about the relationship between nodes. Besides, it is difficult to cost much time and energy to label every node in the large social networks. In this work, we use a factor graph model with partially-labeled data to solve these problems. Our experiments on two data sets (DBLP co-author network, Weibo) with multiple small tasks have demonstrated that our model works much better than the traditional classification algorithms. © Springer-Verlag Berlin Heidelberg 2013. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1007/978-3-642-37453-1_18 | PAKDD |
Field | DocType | Volume |
Factor graph,Data mining,Graph,Data set,Social network,Computer science,Social network analysis,Fringe search,Artificial intelligence,Topic model,Statistical classification,Machine learning | Conference | 7818 LNAI |
Issue | ISSN | Citations |
PART 1 | 16113349 | 4 |
PageRank | References | Authors |
0.42 | 13 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Hua Xu | 1 | 969 | 57.65 |
| Yang Yu-Jiu | 2 | 89 | 19.30 |
| Wang L | 3 | 27 | 2.29 |
| Liu W. | 4 | 209 | 22.42 |