Community-Based Measures for Social Capital

被引:1
|
作者
Spratt, Christopher [1 ]
Hong, Jun [2 ]
McAreavey, Kevin [1 ]
Liu, Weiru [1 ]
机构
[1] Univ Bristol, Bristol BS8 1TH, Avon, England
[2] Univ West England, Bristol BS16 1QY, Avon, England
基金
欧盟地平线“2020”;
关键词
Social capital; Community detection; Social network analysis;
D O I
10.1007/978-3-030-05414-4_26
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Social capital is the value that can be derived from connections between individuals in a social network. The most common forms are bonding and bridging social capital, resulting from connections with similar and diverse groups of individuals respectively. In this paper we propose a novel community-based model for measuring bonding and bridging social capital in a social network. Some previous measures of bonding and bridging capital depend on node attributes, which are often difficult to obtain. Other measures overcome this limitation by relying purely on network structure but are limited to direct connections only for bonding capital and indirect connections only for bridging capital. Our structural measures for bonding and bridging capital are independent of attributes and account for both direct and indirect connections. We experimentally validate our measures on a collaboration network extracted from DBLP, and our results show a strong correlation with standard measures of academic success.
引用
收藏
页码:327 / 338
页数:12
相关论文
共 50 条