Taipei Metro Network Centrality Analysis Using Pagerank Algorithm

In this paper, the node centrality analysis of Taipei metro network is presented by using pagerank algorithm. First, the Taipei metro network is modeled as an undirected graph where each node corresponds to a station and each edge corresponds to a link between stations. Second, the pagerank algorithm is applied to study the importance of node with use of the linkage structure in the undirected graph. Some teleportation mechanisms are provided to bias the random walk and study their performances. Third, a power iteration method which can be regarded as a rational graph filter is used to implement the pagerank algorithm in a distributed manner. The convergence condition of iterative algorithm and polynomial graph filter implementation are also investigated. Finally, experimental results in real-world Taipei metro network are provided to validate the effectiveness of the pagerank algorithm.