Detecting Communities through Network Data

Abstract

Social life clusters into many kinds of groups, a prime topic in the social sciences. Groups often coalesce around common activities, or more generally around social foci (Feld 1981;Kossinets and Watts 2009). Exchanges of information and resources are more frequent within than between groups, which tend to be connected by relatively weaker ties (Granovetter 1973). Some groups may have conflicting relations between them. Conflicts can also exist within groups, but if conflict escalates, groups typically split into opposing factions. In a given population, when neither groups nor group memberships are known beforehand, both can be inferred from social network data. Today, to detect groups or communities in networks, researchers typically use modularity optimization (Fortunato 2010;Reichardt 2009), a method that builds on block modeling (White, Boorman, and Breiger 1976).Network analysts typically assume ties are positive, even though they know not all social relations are positive. Science, for instance, is characterized by cooperation and benign disagreement, but also by epistemic rivalry. In democratic politics, disagreement with 463574A SRXXX10.1177/0003122412463574A merican Sociological ReviewBruggeman et al. 2012

Science of science

“…By applying network science methods to citation networks, researchers are able to identify communities as defined by subsets of publications that frequently cite one another (6). These communities often correspond to groups of authors holding a common position regarding specific issues (7) or working on the same specialized subtopics (8). Recent work focusing on biomedical science has illustrated how the growth of the literature reinforces these communities (9).…”

What a Decade (2006–15) Of Journal Abstracts Can Tell Us about Trends in Ocean and Coastal Sustainability Challenges and Solutions

“…While this exploratory research did not examine the composition of author citation networks (e.g., Wallace et al, 2009;Bruggeman et al, 2012), with ongoing technical advances in software for text citation analysis it may be possible in the not-too-distant future to more seamlessly integrate text mining and bibliometric approaches so as to facilitate co-authorship or bibliometric coupling information in more fine-scale analyses of the academic literature. Empirical research on the formation and composition of epistemic communities is relatively sparse in the environmental science (but see (Sandbrook et al, 2011;Reiners et al, 2015;Rudd, 2015b;Spruijt et al, 2016) as examples of empirical research on epistemic communities or in closely related fields).…”

Robust Science: Passive Smoking and Scientific Collaboration with the Tobacco Industry in the 1970s

“…2 For an elaborated discussion of a quantitative generalized symmetry for the analysis of scientific texts, see the debate in Bearman (2010, 2012) and Bruggeman, Traag, and Uitermark (2012).…”