介数中心性和接近中心性指标, 虽然具有较好的刻画节点重要性的能力, 但计算复杂度太高, 难以在大规模网络上使用. 为了权衡算法的效率和效果, Chen 等人提出了一种基于半局部信息的节点重要性排序方法,简称半局部中心性(semi-local centrality). 首先定义N(w)为节点Vw的两层邻居度，其值等于从 Vw出发 2步内可到达的邻居的数目, 然后定义
其中表示节点 vj的一阶邻居节点的集合. 最终节点 vi的局部中心性定义为
可见, 半局部中心性涉及了节点的四阶邻居信息.
#%%import networkx as nx#定义图的节点和边 nodes=['1','2','3','4','5','6','7','8','9','10','11','12','13','14','15','16','17','18','19','20','21','22','23'] edges=[('1','2',1),('1','3',1),('1','4',1),('1','5',1),('1','6',1),('1','7',1),('1','8',1),('1','9',1),('2','3',1),('3','4',1),('6','10',1),('7','8',1),('8','9',1),('10','11',1),('10','23',1),('11','12',1),('11','21',1),('11','23',1),('12','13',1),('12','14',1),('12','15',1),('13','14',1),('13','15',1),('13','22',1),('14','15',1),('14','23',1),('15','16',1),('16','17',1),('16','18',1),('16','22',1),('17','18',1),('17','20',1),('17','21',1),('18','19',1),('18','22',1),('19','20',1),('19','21',1),('20','21',1),('20','23',1),('22','23',1)] #%%#定义无向图 G = nx.Graph()#往图添加节点和边 G.add_nodes_from(nodes)G.add_weighted_edges_from(edges) #%%N = {}Q = {}CL = {}for node in G.node:node_nei = list(G.neighbors(node))for n_i in node_nei:node_nei = node_nei + list(G.neighbors(n_i))node_nei = list(set(node_nei))N[node] = len(node_nei)-1for node in G.node:node_nei = list(G.neighbors(node))t = 0for n_i in node_nei:t = t + N[n_i]Q[node] = tfor node in G.node:node_nei = list(G.neighbors(node))t = 0for n_i in node_nei:t = t + Q[n_i]CL[node] = tfor node in G.node:print(node,'N-value:',N[node],'Q-value:',Q[node],'CL-value:',CL[node])
结果：
1 N-value: 9 Q-value: 67 CL-value: 1452 N-value: 8 Q-value: 17 CL-value: 923 N-value: 8 Q-value: 25 CL-value: 1014 N-value: 8 Q-value: 17 CL-value: 925 N-value: 8 Q-value: 9 CL-value: 676 N-value: 11 Q-value: 18 CL-value: 1047 N-value: 8 Q-value: 17 CL-value: 928 N-value: 8 Q-value: 25 CL-value: 1019 N-value: 8 Q-value: 17 CL-value: 9210 N-value: 9 Q-value: 37 CL-value: 11111 N-value: 12 Q-value: 41 CL-value: 16612 N-value: 9 Q-value: 38 CL-value: 15713 N-value: 8 Q-value: 39 CL-value: 15714 N-value: 9 Q-value: 40 CL-value: 16615 N-value: 9 Q-value: 37 CL-value: 15616 N-value: 11 Q-value: 39 CL-value: 15817 N-value: 9 Q-value: 39 CL-value: 15818 N-value: 9 Q-value: 40 CL-value: 14819 N-value: 8 Q-value: 28 CL-value: 11920 N-value: 10 Q-value: 40 CL-value: 15821 N-value: 9 Q-value: 39 CL-value: 14822 N-value: 12 Q-value: 42 CL-value: 17023 N-value: 14 Q-value: 52 CL-value: 200
参考：
Duanbing Chen, Linyuan Lü, Ming-Sheng Shang, et al. Identifying influential nodes in complex networks[J]. Physica A: Statistical Mechanics and its Applications, , 391(4):1777-1787.
任晓龙, 吕琳媛. 网络重要节点排序方法综述[J]. 科学通报, (13):1175-1197.
/wyblog/detail/?id=121
