在线社会网络的k-shell结构研究  

Research on k-shell structure of online social networks

作　　者：张艳[1] 张宁[1] 

机构地区：[1]上海理工大学管理学院,上海200093

出　　处：《计算机应用研究》2015年第2期536-538,542,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(70971089);上海市一流学科(系统科学)基金资助项目(XTKX2012);上海市研究生创新基金资助项目(JWCXSL1202/JWCXSL1302)

摘　　要：分析研究了Twitter与You Tube两个在线社会网络的结构。用k-shell(k-壳)分解法对网络分解,并对比分析了它们的入(出)度、入(出)k-shell、以及度与k-shell之间的关系,发现它们之间有较大的差异。You Tube的入(出)度、入(出)k-shell分布均服从幂律分布,而Twitter的分布服从漂移幂律分布、指数截断的幂律分布,但它们的度与k-shell关系基本相同,都未表现出较强的相关性。此外,根据度相关系数的定义还提出k-shell相关性的定义及其计算方法,并用来刻画网络k-shell之间的同(异)配性。This paper presented a large-scale measurement study and analysis of the structure of two online social networks—— Twitter and YouTube. They were decomposed by k-shell decomposition method, then with the analysis and comparison of their in （out）-degree distributions, in （out）-k-shell distributions and the relationships between the degree and k-shell, it shows that there is some significant difference between the two networks. The in （out）-degree distribution, in （out）-k-shell distribution of YouTube both obey power law distribution, however Twitter＇ s obey shifted power law distribution and power law with exponential cutoff respectively. But the relationships between their degree and k-shell were similar, both show no strong correlation. Furthermore, according to the definition of degree correlation coefficient, this paper also put forward the definition of k-shell correlation to characterize the assortativity （disassortativity） between k-shells.

关 键 词：在线社会网络 K-SHELL 度分布 相关系数 

分 类 号：TP393.02[自动化与计算机技术—计算机应用技术]