온라인커뮤니티 사이트에서 하위네트워크의 특성이 정보의 초기 확산에 미치는 영향에 관한 연구

초록

The purpose of this study is to investigate property of sub-network affecting information diffusion at earlier stage. Online community is one of social networks with varying members (nodes) and their relationship(links). Networks in online community have a lot of sub-networks which are set of members sharing similarity in online community network. Their network properties affect to earlier stage of information diffusion. For measuring the property of subnetwork, four network indexes which are path length, clustering coefficient, betweenness centrality, and closeness centrality are used. Path Length is geodesic distance from node i to node j. Clustering coefficient is the proportion of number of triangles in the network-sets of three nodes each of which is connected to each of the others. Betweeness centrality is the proportion of all geodesics between pairs of other nodes that include this node. And, closeness centrality is the number of other nodes divided by the sum of all distance between the node and all others. Four hypotheses tested by network deduction method and statistic analysis. For this study, logdata on C community site was used and analyzed using Pajek 1.91 and, SAS 9.1. The results of this study are as follows. (1) The greater the betweenness or closeness centrality among sub-networks, the greater the diffusion size of sub-network. (2) The shorter the path length or the greater the clustering coefficient, the earlier the average adoption time of sub-network in network. (3) The greater the betweenness centrality or closeness centrality, the earlier the first adoption time.

키워드