Abstract | ||
---|---|---|
Many internet users turn to online knowledge exchange communities to get information they cannot find elsewhere. Question&Answer sites are one of the largest hosts of such communities where users reciprocally answer their questions. Research on expert identification in online communities tries to rank community members by their expertise or to separate experts from non-experts. Until now proposed algorithms for expert identification do not perform well on all datasets. We present an analysis of the structures of topic-induced sub-communities of Question&Answer communities. This analysis aims to provide a basis for expert identification research. The results from the analysis of the network structures explain why common expert identification algorithms are not suitable for all communities. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1145/1880071.1880090 | GROUP |
Keywords | Field | DocType |
community member,common expert identification algorithm,separate expert,knowledge exchange community,answer community,internet user,online community,answer site,expert identification research,expert identification,question answering,social network analysis | World Wide Web,Computer science,Social network analysis,Knowledge management,The Internet,Network structure | Conference |
Citations | PageRank | References |
0 | 0.34 | 8 |
Authors | ||
1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Michael Ovelgönne | 1 | 47 | 5.03 |