中文摘要：

通过对呼叫图、科研合作网络、关键词网络、随机网络4种网络非加权静态特征的计算，得出呼叫图、科研合作网络、关键词网络度和度分布满足幂率分布的无标度网络，也是满足高聚集度、短平均最短路径的小世界网络；同时科研合作网络、呼叫图都是社会网络，而关键词网络是非社会网络。然后引入加权特征，呼叫图和科研合作网络加权聚集度都大于非加权聚集度，即节点倾向于与自己连接次数多的节点连接，这也验证了以上两个网络是社会网络；加权没有显著改变特征的分布，但加权更能精确反映网络的特性；同时由于呼叫图在加权先后节点之间平均权变化比较大，导致呼叫图加权前后特征变化比较显著；网络加权之后聚集度变大、最短路径变小，使得网络的“Small-Word”特性更加明显。

英文摘要：

In this paper, four types network, which are phone call network, scientific co-authorship network, keyword network and random networks are researched. It discusses some metrics to compare the properties among four types of unweighted networks. Then it introduces the weighted metrics to compare the statistical properties, also to find the differences between the unweighted and the weighted networks. We draw a conclusion that the phone call network, co- authorship network, keyword network all display power law shaped degree distribution called scale-free networks, also both have a small value of shortest path and a high clustering coefficient known as the small-world. Phone call network and co-authorship network are social networks, while keyword network isn＇t. The weighted clustering coefficients are larger than unweighted ones and this assortative behavior is in agreement with the evidence that is the two networks are social networks; The analysis of weighted quantities provide a more accurate perspective on the structure of networks, the property of ＂small-world＂ become more prominence because of higher clustering coefficient and smaller value of shortest path compared to unweighted one.