Centrality in social networks based on linear threshold model
Centrality and spread of influence are two basic components in social network analysis. In recent years, the measurement of centrality has attracted the attention of many researchers, and this has led to many new studies on social network analysis and its applications. Older conventional models do not use diffusion of influence extensively to measure centrality. Many studies have been based on the independent cascade model. In this article, the assessment and measurement of centrality based on the linear threshold model in tagged and weighted social networks has been investigated. A method to measure the centrality in order to rank the members of the network is considered, which is named linear threshold rank. To evaluate the proposed method, four social networks have been compared with two measures of centrality and an independent cascade method. The result shows that the proposed method is more suitable for ranking network members and networks. A new centrality measure based on the linear threshold model was presented, which for each member, was considered as the number of members that could spread influence. Compare this metric with three other centrality metrics (Katz centrality, page rank, and independent cascade ranking) based on influence and influence in four real networks, two large networks (one directed, the other undirected) and two small networks (one directed, the other undirected) has been.