Centrality prediction based on K-order Markov chain in Mobile Social Networks

In this paper, we proposed a centrality prediction method based on K-order Markov chains to solve the problem of centrality prediction in Mobile Social Networks (MSNs). First, we use the information entropy to analyze the past and future regularity of the nodes’ centrality in the real mobility traces, and verify that nodes’ centrality is predictable. Then, using the historical information of the center of the node, the state probability matrix is constructed to predict the future central value of the node. At last, through the analysis of the error between real value and predicted value, we evaluate the performance of the proposed prediction methods. The results show that, when the order number is K = 2, compared with other existing four time-order-based centrality prediction methods, the proposed centrality prediction method based on K-order Markov chain performs much better, not only in the MIT Reality trace, but also in the Infocom 06 traces.