Using Topical Networks to Detect Editor Communities in Wikipedias
Year | 2019 |
---|
The collaboration of Wikipedia editors is well researched, covered by scientific works of many different fields. There is a growing interest to implement recommender systems that guide inexperienced editors to projects which fit their interests in certain topical domains. Although there have been numerous studies focusing on editing behavior in Wikipedia the role of topical domains in this regard is still unclear. In particular, topical aspects of co-authorship are generally neglected. In this paper, we want to determine by which criteria editors usually choose articles they want to contribute to. We analyzed three different language editions of Wikipedia (Vietnamese, Hebrew, and Serbo-Croatian) by building social networks and running community detection algorithms on them, i.e. editors are grouped based on their shared involvement in Wikipedia articles using social network analysis techniques. Then, we related this to the topical domains of these articles based on Wikipedia’s user defined category network. Our results demonstrated that communities in Wikipedia tend to edit articles with a higher than average topical relatedness. But the significance and quality of these results vary considerably in the different language versions of Wikipedia. Topical relations between contributors and articles are a complex matter and influenced by a number of different factors, e.g. by culture.
Details
2019 Sixth International Conference on Social Networks Analysis, Management and Security (SNAMS),DOI: 10.1109/SNAMS.2019.8931865
Authors
- Michael Kretschmer
- Bernhard Göschlberger
- Ralf Klamma
Presented at
Sixth International Conference on Social Networks Analysis, Management and Security (SNAMS), 2019 .
Published in
Sixth International Conference on Social Networks Analysis, Management and Security (SNAMS) .