E-Mail as Spectroscopy: Automated Discovery of Community Structure within Organizations

By Joshua R. Tyler, Dennis M. Wilkinson, Bernardo A. Huberman. The Information Society, 2003

665 citations recorded by the source corpus.

We describe a method for the automatic identification of communities of practice from e-mail logs within an organization. We use a betweenness centrality algorithm that can rapidly find communities within a graph representing information flows. We apply this algorithm to an initial e-mail corpus of nearly 1 million messages collected over a 2-month span, and show that the method is effective at identifying true communities, both formal and informal, within these scale-free graphs. This approach also enables the identification of leadership roles within the communities. These studies are complemented by a qualitative evaluation of the results in the field.

Subfields: Computer Vision and Pattern Recognition, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/1b65e308316cda760a3794eb65e0aeeae27f9cfb

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Statistical and Nonlinear Physics

Citations in the corpus

POLYPHONET: an advanced social network extraction system from the web (2006, background): “In contrast, automatic detection of relations is also possible from various sources of information such as e-mail archives, schedule data, and Web citation information [1, 44, 34].”
Why social networks are different from other types of networks. (2003, background): “Recent work [9, 29, 30, 31, 32] suggests a possible candidate theory, that social networks contain groups, or “community structure” [41].”
Finding community structure in very large networks. (2004, background): “I. INTRODUCTION Many systems of current interest to the scientific community can usefully be represented as networks [1, 2, 3, 4].”
Characterization of complex networks: A survey of measurements (2005, methodology): “algorithm runs in worst-case timeO(M2N) on networks with M edges and N vertices. In order to overcome this limitation, some improvements in the algorithm were proposed including the Tyler’s algorithm [135], which introduced a stochastic element to the method, restricting the calculation of the betwe”
Community detection in social networks (2016, methodology): “The Girvan Newman (GN) algorithm has been enhanced by many authors and applied to various networks 21 – 28 .”
Community detection in Social Media (2012, positive): “They also use the efficient sample-based version of the algorithm (Tyler et al. 2003) and observe that the derived communities are the same.”
Extremal Properties of Random Structures (2003, positive): “. Alani et al. [4] recently introduced a semi-automated utility that uses a simple algorithm to identify nearest neighbors to one individual within a university department. The method of Tyler et al. [46] uses email data to construct a network of correspondences, and then discovers the communities b”
Fast algorithm for detecting community structure in networks. (2003, background): “…identify such structure with some sensitivity, and this algorithm has been employed by a number of authors in the study of such diverse systems as networks of email messages, social networks of animals, collaborations of jazz musicians, metabolic networks, and gene networks [5, 6, 7, 8, 9, 10, 11].”
Community Structure in Graphs (2007, methodology): “Tyler, Wilkinson and Huberman proposed a modi-ﬁcation of the Girvan-Newman algorithm, to improve the speed of the calculation [40].”
A framework for analysis of dynamic social networks (2006, background): “Finding patterns of social interaction within a population has wide-ranging applications including: disease modeling [14, 22], cultural and information transmission [4, 6, 9, 18, 30, 32], intelligence and surveillance [4, 21, 26], business management [5, 8, 28, 29], conservation biology and behavior”
A review of modularization techniques in artificial neural networks (2019, methodology): “In more recent approaches (Newman, 2004; Tyler et al., 2005; Radicchi et al., 2004), a modularity measure (Equation 1) was used to guide the detection process towards the best division.”
Rhythms of social interaction: messaging within a massive online network (2006, methodology): “Figure 1.4 (color) shows messaging activity by hour in a corporate email network, using the same dataset as used in [17].”