A Research for the Centrality of Article Edit Collective in Wikipedia

Aiming at the problems of the centrality of article edit collective in wikipedia, under the direction of the idea of networked data mining, featured articles in wikipedia were analyzed by text processing to find the difference of sentences between adjacent versions and identify the edit interaction connection between editors, then the article edit interaction networks were constructed, where the node is editor and the link is the edit interaction connection between editors, then degree, between ness and closeness and topology potential were used to analyze empirically the local centrality of article edit interaction networks. Results show that the cumulative distributions for degree, between ness and topology potential of nodes follow shifted power law distribution, closeness follows normal distribution, and there are many nodes with small degree and between ness but big closeness, there are few nodes with big degree, between ness and closeness. There isn´t an absolute center in the networks. However the edit collective have strong heterogeneity and local community structure and topology potential can synthetically characterize the centrality of nodes. The method can effectively find the central nodes in the networks and the research deepens the knowledge of the characteristic of collective edit interaction and collective intelligence.