Source: Think Links
Sunbelt is the annual meeting for Social Network Analysis researchers. It’s been going on since 1981 (a couple of years before analyzing twitter graphs became hip) and this year it’s being held in Tampa. Two of my colleagues-Julie Birkholz and Shenghui Wang- are attending and presenting some joint work. The abstracts are below. If you’re at Sunbelt be sure to check out their presentations and have a chat.
At a higher level, I think both pieces of work emphasize the importance of using the combination of rich representations of the data underlying networks along with dynamic network analysis. Networks provide a powerful abstraction mechanism but it’s important to be able to situate that abstraction in a rich context. The techniques we are both developing and applying are steps along the way towards enabling these more “situated” network.
Dynamics Of Scientific Collaboration Networks
Groenewegen, Peter; Birkholz, Julie M.; van der Bunt, Gerhard; Groth, Paul
Evolution of scientific research can be considered as a dynamic network of collaborative relations between researchers. Collaboration in science leads to social networks in which authors can gain prominence through research (knowledge production), access to highly regarded field members, or network positions in the collaborative network. While a central position in network terms can be considered a measure of prominence, the same holds for citation scores. Causal evidence on a central position in the network corresponding to prominence in other dimensions such as the number of citations remains open. In this paper collaborative patterns, research interests and citation counts of co‐authoring scientists will be analyzed using SIENA to establish whether network processes, community or interest strategies lead to status in a scientific fields, or vice versa does status lead to collaboration. Results from an analysis of a subfield of computer science will be presented.
Multilevel Longitudinal Analysis For Studying Influence Between Co‐evolving Social And Content Networks
Wang, Shenghui; Groth, Paul; Kleinnijenhuis, Jan; Oegema, Dirk A
The Social Semantic Web has begun to provide connections between users within social networks and the content they produce across the whole of the Social Web. Thus, the Social Semantic Web provides a basis to analyze both the communication behavior of users together with the content of their communication. However, there is little research combining the tools to study communication behaviour and communication content, namely, social network analysis and content analysis. Furthermore, there is even less work addressing the longitudinal characteristics of such a combination. This paper proposes to take into account both the social networks and the communication content networks. We present a general framework for measuring the dynamic bi‐directional influence between co‐evolving social and content networks. We focus on the twofold research question: how previous communication content and previous network structure affect (1) the current communication content and (2) the current network structure. Multilevel time‐series regression models are used to model the influence between variables derived from social networks and content networks. The effects are studied at the group level as well as the level of individual actors. We apply this framework in two use‐cases: online forum discussions and conference publications. By analysing the dynamics involving both social networks and content networks, we obtain a new perspective towards the connection of social behaviour in the social web and the traditional content analysis.