On the use of external information in social network analysis
| 详细信息 | 推荐本文 | |
- 作者：Giuseppe Giordano (1) ggiordan@unisa.it
Maria Prosperina Vitale (1) - 关键词：Attribute data – ; Clustering – ; Positional analysis – ; Scientific collaboration network – ; Two ; mode network
- 刊名：Advances in Data Analysis and Classification
- 出版时间：July 2011
- 出版年：2011
- 期刊代码：pc562_1862-5347
- 类别：mat
- 卷：5
- 期：2
- 页码：95-112
- 数据来源：sp
摘要
Network analysis focuses on links among interacting units (actors). Interactions are often derived from the presence of actors at events or activities (two-mode network) and this information is coded and arranged in a typical affiliation matrix. In addition to the relational data, interest may focus on external information gathered on both actors and events. Our aim is to explore the effect of external information on the formation of ties by setting a strategy able to decompose the original affiliation matrix by linear combinations of data vectors representing external information with a suitable matrix of coefficients. This allows to obtain peculiar relational data matrices that include the effect of external information. The derived adjacency matrices can then be analyzed from the network analysis perspective. In particular, we look for groups of structurally equivalent actors obtained through clustering methods. Illustrative examples and a real dataset in the framework of scientific collaboration will give a major insight into the proposed strategy.