The Catalyst Conference was July 27-31 in San Diego which gathered many people to discuss new technologies for enterprise computing. On Wednesday, July 29 at 11:15 a.m. I gave a presentation about the use of social network analysis within the enterprise. 2009…
A new book E-Research: Transformation in Scholarly Practice edited by Nicholas W. Jankowski on the ways social science research is being changed by the rise of social media has just been released by Routledge. My colleagues and I contributed a chapter on the ways that information visualization of social media is a useful technique to identify research questions and discover answers about the nature of human association when mediated by computation. The volume contains work from an all-star line-up of researchers who address the opportunities and challenges of performing research with computer-mediated data about social life.
The blurb about the book describes it as:
“No less than a revolutionary transformation of the research enterprise is underway. This transformation extends beyond the natural sciences, where ‘e-research’ has become the modus operandi, and is penetrating the social sciences and humanities, sometimes with differences in accent and label. Many suggest that the very essence of scholarship in these areas is changing. The everyday procedures and practices of traditional forms of scholarship are affected by these and other features of e-research. This volume, which features renowned scholars from across the globe who are active in the social sciences and humanities, provides critical reflection on the overall emergence of e-research, particularly on its adoption and adaptation by the social sciences and humanities.”
Our chapter is “A Picture is Worth a Thousand Questions: Visualization Techniques for Social Science Discovery in Computational Spaces”, co-authored by Howard T. Welser, Thomas Lento, Marc Smith, Eric Gleave and Itai Himelboim. In it, we describe the ways that using information visualizations of social media data sets is a useful way of discovering insights, patterns, and clusters. We illustrate the paper with several examples of social media information visualizations that display the range of behavior among contributors to social media spaces.
Here is the table of contents for the volume:
My colleagues Derek Hansen and Ben Shneiderman (University of Maryland) and I have just finished the second version of our tutorial/manual for the NodeXL social network analysis toolkit for Excel.
The latest version of the tutorial Analyzing Social Media Networks: Learning by Doing with NodeXL is now available from the University of Maryland Center for the Advanced Study of Communities and Information (CASCI) web site. We will use this version of the document in our upcoming tutorial at the Communities and Technologies conference at Penn State University on June 24th.
We plan to continue to expand the tutorial to include a step-by-step guide to the analysis of several major social media sites like Twitter, Facebook, Wikipedia, YouTube, delicious, and flickr as well as personal stores of social media like your own email (if it is stored in a Windows Search Index found on most Windows desktops). Our goal is to create an easy-to-follow guide to network theory for people who new to the field or who do not want to develop programming skills to perform network analysis. We are focused on social media as a data source for social media although other examples are included, like the United States Senate voting network that reveals interesting patterns in the connections created when votes are cast. Using 2007 data it reveals which Senators are most likely to change party affiliation.
Your comments, corrections, and suggestions for improving the document are welcome.
Instructors interested in teaching classes about social networks are welcome to make use of both the NodeXL toolkit and the document to guide students through the core concepts of social network theory.
Here is the table of contents:
[flickrset id="72157617551332795" thumbnail="square" overlay="true" size="medium"] A few weeks ago I attended a meeting at the University of Maryland in College Park of a working group proposing a new "National Initiative for Social Participation". The meeting brought together people from the…
10.30am A categorical model for discovering latent structure in social annotations (Said Kashoob)
Given a collection of web objects, users and tags, can we model the underlying tag generation process?
-Discover implict communities of interest?
-Categories of related tags?
-For given category, id most relevant objs for category
Initial thoughts: content-based topic modeling (Latent Dirichlet Allocation, LSA). Recent work applying LDA models to tags (Wu 2006, Zhou 2008)