A. An et al. (Eds.): AMT 2010, LNCS 6335, pp. 47–62, 2010.
© Springer-Verlag Berlin Heidelberg 2010
Visualizing Threaded Conversation Networks:
Mining Message Boards and Email Lists
for Actionable Insights*
Derek L. Hansen1, Ben Shneiderman2, and Marc Smith3
1
College of Information Studies & Center for the Advanced Study of Communities
and Information, University of Maryland, College Park, Maryland, USA
dlhansen@umd.edu
2
Dept. Of Computer Science & Human-Computer Interaction Lab,
University of Maryland, College Park, Maryland, USA
ben@cs.umd.edu
3
Connected Action Consulting Group, Silicon Valley, California, USA
marc@connectedaction.net
Abstract. Analyzing complex online relationships is a difficult job, but new in-
formation visualization tools are enabling a wider range of users to make action-
able insights from the growing volume of online data. This paper describes the
challenges and methods for conducting analyses of threaded conversations such
as found in enterprise message boards, email lists, and forums. After defining
threaded conversation, we characterize the types of networks that can be ex-
tracted from them. We then provide 3 mini case studies to illustrate how action-
able insights for community managers can be gained by applying the network
analysis metrics and visualizations available in the free, open source NodeXL
tool, which is a powerful, yet easy-to-use tool embedded in Excel 2007/2010.
1 Introduction
Threads are the things that hold the net together. Since the inception of the Internet
most virtual communities have relied on asynchronous threaded conversation plat-
forms as a main channel of communication. Usenet newsgroups, email lists, web
boards, and discussion forums all contain collections of messages in reply to one
another. The natural conversation style supported by the basic post-and-reply threaded
message structure has proven enormously versatile, serving communities ranging
widely in focus and goals. Cancer survivors and those seeking technical support or
religious guidance are as likely to use a threaded discussion as a corporate workgroup.
Modern incarnations of threaded conversation are embedded in social networking site
wall posts, blog comments, Google Wave threads, YouTube or Flickr comments, and
Twitter ‘reply to’ (RT) tweets. Traditional forums now include profile pages, partici-
pation statistics, reputation systems, and private messaging.

*
This paper is a revised version of a chapter from “Analyzing Social Media Networks with
NodeXL: Insights from a Connected World” by Hansen, Shneiderman, and Smith to be pub-
lished by Morgan Kaufmann Publishers in Fall 2010.

48 D.L. Hansen, B. Shneiderman, and M. Smith
Despite the differences in types of threaded conversation, the common structure
lends itself well to network analysis, due to its easily identifiable reply structure that
captures communication patterns between people. Unfortunately, most threaded con-
versation systems do not make this networked data easily accessible. The majority of
threaded message content is not easily accessible due to the number of different soft-
ware platforms used and the fact that many groups only make content accessible to
subscribed members. Many threaded message systems do report participation statis-
tics and ratings (e.g., top 10 contributors), which are important metrics but fail to
capture the social connections between members – a critical component of virtual
communities and corporate communities of practice.
This paper considers how to analyze threaded conversations from a network per-
spective. We begin by defining threaded conversation and characterizing some of the
most important networks that can be created from threaded conversation. We then
include several brief case studies that demonstrate the value of taking a network ap-
proach. The major contribution is to demonstrate novel analysis and visualization
approaches that provide users with powerful methods for extracting actionable in-
sights. We rely upon a novel, open source network analysis tool called NodeXL
(www.codeplex.com/nodexl), which enables a wider range of analysts to make dis-
coveries and visual presentations that previously required a higher degree of technical
skills. These analysts can apply their rich domain knowledge and understanding of
social and organizational structures to handle larger datasets and make appropriate
business decisions.
2 Definition and Structure of Threaded Conversation
Threaded conversation is a commonly used design theme that enables online discus-
sion between multiple participants using the ubiquitous post-reply-reply structure. It
shows up in many forms from email lists to web discussion forums to photo sharing
and customer review sites. The key properties of threaded conversation were enumer-
ated in Resnick, et al. [1] and are listed here with some modification:
• Topics. A set of topics, groups, or spaces, sometimes hierarchically organized to
aid users in discovering interesting groups to “join.” Topics or groups are persis-
tent, though their contents may change over time. Fig. 1 includes two topics:
TOPIC 1: Social Media and TOPIC 2: NodeXL.
• Threads. Within each topic or group, there are top-level messages and responses
to those messages. Sometimes further nesting – responses to responses – is permit-
ted. The top-level message and the entire tree of responses to it are called a thread.
In Fig. 1, there are 5 unique threads. Thread A includes only 2 messages, while
Thread B includes 6 messages. Thread D includes only a single message.
• Single Authored. Each message contributed to a thread is authored by a single
user. Typically, the person’s username or email address is shown alongside the
post so people know who is talking. In Fig. 1, the author of each message and the
time of their post are indicated. Users may post to multiple threads (e.g., Beth) or
multiple times within a thread (e.g., Cathy).
• Permanence. In many threaded conversations including email lists and Usenet,
once a message has been posted it cannot be re-written or edited. A new message