Analyzing Social Media Networks with NodeXL: Insights from a Connected World

Post date: Oct 01, 2011 10:1:35 AM

For the last dozen years, networks have attracted the attention of many researchers and practitioners. Some simple network models [1,2] have achieved significant popularity beyond Andy Warhol’s 15 minutes of fame. Such models have spurred the development of a whole new field of research that combines graph theory, computer science, economics, and other social sciences. The generated momentum has led to the recent publication of some outstanding textbooks on networks [3,4,5].

More than a classical textbook, this book is a well-written educational handbook that describes how we can apply part of the so-called science of networks to the analysis of medium-size networks. The authors impose the size constraint because of the book’s reliance on visual inspection and interactive visualization techniques (sometimes referred to as visual analytics). Since the available space on a screen is severely limited, one cannot expect to analyze a billion-node network using the tool described in this book.

The book also provides a nice overview of the field of social media networks. (Social media corresponds to Web 2.0 technologies.)

A surprising feature of this book, especially when compared with its more formal counterparts, is that the authors avoid any kind of mathematical formalization of the topics they discuss. They seem to have followed Stephen Hawking’s advice: “Someone told me that each equation I included in the book would halve the sales” [6]. In fact, many readers will be grateful to learn that the authors do not include a single mathematical equation, not even to define important network metrics from an intuitive point of view. (Stephen Hawking did include a single equation: Einstein’s famous E=mc2.) However, some chapters include “Advanced Topics” sidebars that describe how to get the most out of Excel when using the NodeXL tool.

NodeXL is a social network analysis open-source plug-in for Excel 2007/2010. It lowers the barrier of entry that is typical of many other network analysis tools; anyone with some experience using spreadsheets can use it. NodeXL computes some of the best-known network metrics, and, with the help of some Excel features, makes network filtering and visualization a breeze. It also implements different network layout algorithms and clustering techniques to discover natural groups of nodes within a complex network. The book only mentions one clustering technique, but more alternatives are available by downloading the latest version of the software from the NodeXL homepage (http://nodexl.codeplex.com/).

The introductory part of the book goes through some of the history and core concepts of social media and social network analysis, always with an eye to social network visualization and for a general audience--it omits technical details. Later, it proposes a framework for categorizing different collaboration technologies and describes 11 categories as examples.

The second part of the book features a variety of tool details. It presents an eight-hour tutorial that surveys NodeXL’s main features.

The bulk of the book, however, is the third part, which contains eight chapters on as many different case studies. These chapters, which are often by international researchers on social media, focus on different kinds of social networks and follow a more or less standard structure. They cover email, threaded communications, Twitter, Facebook, organizational World Wide Web (WWW) networks, Flickr, YouTube, and wikis.

The chapters describe the particularities of each technology and present information on the types of networks one can analyze, from Facebook ego networks to the interplay between content and community networks on Flickr or YouTube (that is, the content provided by tag clouds and video networks versus the structure provided by user networks, where you can explore each user’s followers and/or friends). Each chapter concludes with some observations for practitioners, an agenda for researchers, and a good selection of relevant bibliographic references for each case study. Ideally, these analyses should provide some useful insights and a good idea of what types of questions one can ask (and answer using social network analysis) about individuals, groups, temporal evolution, and structural patterns in social media networks.

In summary, Hansen, Shneiderman, and Smith, along with their collaborators, have written a readable introduction to the field of social media network analysis. Furthermore, the book is a nice tutorial on an interesting tool that readers can experiment with on their own. For example, the readers can simply use the Facebook application provided by Bernie Hogan, one of the book’s collaborators, to analyze the ego networks that they know best: their own network of friends, family, and acquaintances. This book offers a sure way to understand some of the basic concepts of network analysis.

Reviewer: Fernando Berzal

Review #: CR138913 (1111-1143)

1)

2)

3)

4)

5)

6)

Watts, D.J.; Strogatz, S.H. Collective dynamics of “small-world” networks. Nature393, 6684(1998), 440–442.

Barabási, A.-L.; Albert, R. Emergence of scaling in random networks. Science 286, 5439(1999), 509–512.

Newman, M.E.J. Networks: An Introduction. Oxford University Press, Oxford, UK, 2010.

Easley, D.; Kleinberg, J. Networks, Crowds, and Markets: Reasoning about a Highly Connected World. Cambridge University Press, New York, NY, 2010.

Jackson, M.O. Social and Economic Networks. Princeton University Press, Princeton, NJ, 2008.

Hawking, S.W. A brief history of time: from the big bang to black holes. Bantam Books, New York, NY, 1988.

También disponible en