Analyzing Email Communications: An Ego-Centric Approach

As a quick scan through prior blogs will show, throughout this year we have been exploring the application of social network visualization software to email communications. Our interest has been two-fold: finding tools to support those working in legal and regulatory environments who need to examine large numbers of emails for answers to “Who, What, Where, When and Who Knew” kinds of questions and secondly, to see if this approach might provide behavioral psychologists with tools to identify and/or objectively measure, communications issues in workplace teams. In many workplace situations, email has become the primary communication mechanism whether through cultural factors (as with many IT teams) or because of distance (with geographically dispersed teams). At the same time communication issues are cited as one of the primary reasons why projects fail. It seemed to us that tools for analyzing the flow of email communications in a team might help identify team members who are outside the group, or who have significantly fewer interactions with key individuals in the team, thereby enabling remedial action to be taken.

Software we have looked at so far includes: Gephi – useful for large data sets – and NodeXL – useful for analyzing smaller groups of individuals with great options for customizing the appearance of the graphs e.g. color coding particular attributes or clusters and easy to use. Data feeds into both are organized basically as edge lists and node lists with Gephi requiring XML formatting and NodeXL spreadsheet or csv lists. (Note: in an email environment, a node is an individual – represented by either an email address or a name and an edge is the communication between two individuals with the volume of communications represented by a weight measure). The visualizations produced look at communication and clustering from a birds-eye view across the entire data set.

UCINET takes a somewhat different approach. UCINET is a social network analysis program developed by university researchers at the University of Kentucky and distributed by Analytic Technologies (see www.analytictech.com/ucinet/). There is a free trial version and relatively low cost options for students, researchers and single users.

Unlike NodeXL or Gephi, UCINET is not a complete visualization package but only the analytic engine. It is, however, integrated with a freeware program called NETDRAW. Since both are included in the download package, installation is straightforward. We did find in practice though that the package behaves like a set of separate tools operating on a common data set compared with the more integrated environments of NodeXL or Gephi. Another difference is that UCINET works on matrices not edge/node lists. Fortunately, it has an import function which accepts a standard edge list (e.g. person1, person2, weight) in excel format. The import function then converts this into a matrix for analysis and visualization.

Our test data set is the same as before: an anonymized set of email communications. For this investigation we started with a small subset of 368 nodes and 1223 edges.

NETDRAW visualization of entire email network

While NETDRAW is by no means as sophisticated as the graphical packages in Gephi or even NodeXL, where the UCINET/NETDRAW package came into its own is in its ability to hone in easily on a selected set of individuals. A checklist menu of nodes appears on the right hand side of the graph and altering the selections immediately redraws the graph showing only those individuals and their connections. We think this is very helpful when drilling down to investigate the interactions between a particular group of people.

Another great feature of UCINET/NETDRAW is its ability to visualize interactions from an “ego” perspective. By selecting an initial “ego”, the software identifies all the individuals in communication with the selected individual and produces a subgraph of communications between them. For example, simply selecting “Carmela Soprano” produced the following subgraph.

"Carmelo Soprano" Ego Network Graph

NETDRAW can be configured to represent the volume of communications as the size of the link:

Network Graph with Link Width Representing Communication Volume

Or with the volume shown in a link label:

Network Graph with Link Label Showing Communication Volume

UCINET offers a range of node centrality measures including Closeness, Betweenness, Degee and Eigenvector. (For information about what these measures represent, see previous blogs or go to: http://en.wikipedia.org/wiki/Betweenness_centrality#Eigenvector_centrality). Once the measures are calculated, nodes can be colorized to represent one of the selected measures. For example the nodes on the sub-graph below have been colorized to represent the value of the Indegree attribute.
It is also possible to filter based on a particular measure. The graph below shows the entire set filtered to show only nodes with high Eignvector counts (a measure of the importance of the individual in the network).

Network filtered by Eigenvector Measure (to show 'Important' individuals only)

UCINET/NETDRAW also has a number of algorithms for analyzing subgroups. For example, in the subgraph below (an “ego” network for Tom Hagen), it has identified 3 factions – represented by the three different colors: red, blue, black.

Graph identifying Factions within a Subgroup

An analysis of cliques in the entire set identified 60 separate groups shown in the graph below.

Graph showing the 60 cliques identified in the data set

What we liked about UCINET/NETDRAW is the ease with which we could explore the involvement of particular individuals in the network using the ego feature combined with the filtering and attribute based node coloring. We also liked the wide range of analysis options which included not only the standard centrality measures but also various clustering algorithms and analyses of cliques and subgroups. While more extensive documentation would have been helpful, (although we do appreciate that this was initially developed as a research tool), we did appreciate that whatever we did to it, it never crashed and managed to catch any errors gracefully.

Visualizing Email Communications using NodeXL

Email has become an integral part of communication in both the business and personal spheres. Given its centrality, it is surprising how few tools are generally available for analyzing it outside specialist areas such as Early Case Assessment tools within the litigation area: Xobni being a notable exception at the individual level. However, the rise of social network analysis, and the tools that support it, may change that. Graph theory is remarkably neutral as to whether it is applied to Facebook Friend networks or email communications within a Sales and Marketing division.

In a previous post, we reported on using Gephi – an open source tool for graphing social networks – to visualize email communications. In this post, we look at using NodeXL for the same purpose. We used the same email data set before – the ‘Godfather Sample’ – in which an original email data set was processed to extract the metadata (e.g. sender, recipient, date sent, subject) and subsequently anonymized using fictional names.

NodeXL is a free and open source template for Microsoft Excel 2007 that provides a range of basic network analysis and visualization features intended for use on modest-sized networks of several thousand nodes/vertices. It is targeted at non-programmers and builds upon the familiar concepts and features within Excel. Information about the network, e.g. node data and edge lists, is all contained within worksheets.

Data can be simply loaded by cutting and pasting an edge list from another Excel worksheet but there are also a wide range of other options including the ability to import network data from Twitter (Search and User networks), YouTube and Flickr and from files in GraphML, Pajek and UCINET Full Matrix DL file formats. There is also an option to import directly from an Email PST file which we will discuss a following post. In addition to the basics of an edge list, attribute information can be associated with each edge and node. In our “Godfather” email sample, we added a weighting for communication strength (i.e. the number of emails between the two individuals) to each edge and the affiliation with the Corleone family to each node.

Once an edge list has been added, the vertices/node list is automatically created and a variety of graphical representations can be produced depending on the layout option selected, (Fruchterman Riengold is the default but Harel-Koren Fast Multiscale as well as Grid, Polar, Sugiyama and Sine Wave options are also available), and by mapping data attributes to the visual properties of nodes and vertices. For example, in the graph shown below, nodes were color coded and sized with respect to the individual’s connections with the Corleone family: blue for Corleone family members, green for Corleone allies, orange for Corleone enemies and Pink for individuals with no known associations with the family.

The width of the edges/links was then set to vary in relation to the degree of communication between the two nodes i.e. the number of emails sent between the two individuals concerned.

Labels can be added to both nodes and links showing either information about the node/link or its attributes, as required.

Different graph layout options are available which may be used to generate alternative perspectives and/or easier to view graphs.

Harel-Koren Layout

Circle Layout

Because even a small network can generate a complex, dense graph, NodeXL has a wide range of options for filtering and hiding parts of the graph, the better to elucidate others. The visibility of an edge/vertex for example, can be linked to a particular attribute e.g. degree of closeness. We found the dynamic filters particularly useful for rapidly focusing on areas of interest without altering the properties of the graph themselves. For example, in the following screenshot we are showing only those links where the number of emails between the parties is greater than 40. This allows us to focus on individuals who have been emailing each other more frequently than the average.

In addition to graphical display, NodeXL can be used to calculate key network metrics including: Degree (the number of links on a node and a reflection of the number of relationships an individual has with other members of the network) with In-Degree and Out-Degree options for directed graphs, Betweenness Centrality (the extent to which a node lies between other nodes in the network and a reflection of the number of people an individual is connecting to indirectly), Closeness Centrality (a measure of the degree to which a node is near all other nodes in a network and reflects the ability of an individual to access information through the "grapevine" of network members) and Eigenvector Centrality (a measure of the importance of an individual in the network). In an analysis of email communications, these can be used to identify the degree of connectedness between individuals and their relative importance in the communication flow.

For example, in our Godfather sample, we have sized the nodes in the graph below by their Degree Centrality. While Vito Corleone is, as expected, shown to be highly connected, Ritchie Martin – an individual not thought to have business associations with the Corleone family, is shown to be more connected than supposed.

Node Sized by Degree Centrality

When we look at the same data from the perspective of betweenness, we see that Vito, Connie and Ritchie all have a high degree of indirect connections.

Nodes Sized by Betweenness Centrality

And the Eignevector Centrality measure confirms Vito Corleone's signficance in the network as well as Connie's, two "allies" - Hyman Roth and Salvatore Tessio and two individuals  Ritchie Martin.

Nodes Sized by Eigenvector Centrality

Last but not least, it is also possible to use NodeXL to visualize clusters of nodes to show or identify subgroups within a network. Clusters can be added manually or generated automatically. Manually creating clusters requires first assigning nodes to an attribute or group membership and then determining the color and shape of the nodes for each subgroup/cluster. In our GodFather example, we used “Family” affiliation to create clusters within the network but equally one could use organization/company, country, language, date etc.

"Family Affiliation" Clusters Coded by Node Color

Selected Cluster (Corleone Affiliates)

NodeXL will also generate clusters automatically using a clustering algorithm developed specifically for large scale social network analysis which works by aggregating closely interconnected groups of nodes. The results for the Godfather sample are shown below. We did not find the automated clustering helpful but this is probably a reflection of the relatively small size of the sample.

In the next post, we will look at importing email data directly into NodeXL and compare approaches based on analyzing processed vs unprocessed email data.

Larger Email Network Visualization

To download NodeXL, go to http://nodexl.codeplex.com//. We would also recommend working though the NodeXL tutorial which can be downloaded from: http://casci.umd.edu/images/4/46/NodeXL_tutorial_draft.pdf


A top level overview of social network analysis and the basic concepts behind graph metrics can be found on Wikipedia e.g. http://en.wikipedia.org/wiki/Social_network and http://en.wikipedia.org/wiki/Betweenness_centrality#Eigenvector_centrality