基于改进模糊聚类算法的地铁站点重要性分析

为了研究地铁网络的节点重要性,研究一种改进模糊聚类算法:将地铁网络抽象为无向无权复杂网络,选取节点介数、接近中心性和特征向量中心性作为节点重要性的评价指标,利用k-means和熵权法改进的FCM模糊聚类算法,对地铁网络节点的重要性进行软化分.采用误判率交叉估计法对模糊聚类结果进行检验.实证表明,相比传统FCM算法,改进FCM算法目标函数收敛次数降低了58.06％,目标函数值降低了5.33％,误判率为0,更适用于地铁站点的重要性分析.

Importance Analysis of Metro Station Based on Improved Fuzzy Clustering Algorithm

In order to study the importance of metro network nodes, an improved fuzzy clustering algorithm is studied.Metro network is abstracted as an undirected and unweighted complex network, node betweenness, proximity centrality and centrality of eigenvector are selected as the evaluation indexes of node importance, and FCM fuzzy clustering algorithm improved by k-means and entropy weight method is used to soften the importance of metro network nodes.The cross estimation method of misjudgment rate is used to test the clustering results. The empirical results show that, compared with the traditional FCM algorithm, the convergence times of the objective function of the improved FCM algorithm is reduced by 58.06%, the value of the objective function is reduced by 5.33%, and the misjudgment rate is reduced to 0, which is more suitable for the importance analysis of metro stations.