People talk about the products and services the use, love or hate all the time in social media. These conversations can be better understood through perspective of social network analysis. Network theory views the world as a web of connected people. Network analysis builds measures and visualizations of collections of connections to reveal the key people, groups and issues in these conversations. Using social media network maps and reports the confusing landscape of tweets and posts comes into focus. Information visualizations of the virtual crowds of people gathered around every brand, product, event, or service highlights the range of variation in the shape of these crowds. Six different patterns have been identified so far, allowing social media managers to recognize the nature of the brand network they have and the nature of the network they want to have. Network measures are useful as KPIs for tracking not just the size and volume of a social media stream, but also the shape and structure of the pattern of connections. The six patterns: divided, unified, fragmented, clustered, and in and out hub and spoke, are a useful guide to strategic engagement in social media.
The Social Media Research Foundation team has innovated at multiple levels: organizationally we are a modern, virtual, distributed group of collaborators. Technically, we have focused our project on ease of use and automation rather than scale and sophistication, our users are not programmers. We have implemented many innovative network analysis and visualization techniques because we have been driven by a need to serve a diverse user population. The contributors to the project are themselves from a diverse range of disciplinary backgrounds, making it easier to shape the tool for the broadest audience.
The talk will focus on the easy to follow steps needed to create social media network maps and reports automatically from services like Twitter, Facebook, YouTube, Flickr, email, blogs, wikis, and the WWW. Here is a sample network map of the term #bigdataprivacy:
The graph represents a network of 248 Twitter users whose recent tweets contained “#bigdataprivacy”, or who were replied to or mentioned in those tweets. The tweets in the network were tweeted over the 6-day, 10-hour, 29-minute period from Tuesday, 25 February 2014 at 14:36 UTC to Tuesday, 04 March 2014 at 01:06 UTC. There is an edge for each “replies-to” relationship in a tweet. There is an edge for each “mentions” relationship in a tweet. There is a self-loop edge for each tweet that is not a “replies-to” or “mentions”.
The graph’s vertices were grouped by cluster using the Clauset-Newman-Moore cluster algorithm.
The graph was laid out using the Harel-Koren Fast Multiscale layout algorithm.
The edge colors are based on edge weight values. The edge widths are based on edge weight values. The edge opacities are based on edge weight values. The vertex sizes are based on followers values. The vertex opacities are based on followers values.
Top 10 Vertices, Ranked by Betweenness Centrality:
@whitehouseostp, @mit, @mit_csail, @steve_lockstep, @aureliepols, @dbarthjones, @digiphile, @stannenb, @djweitzner, @mikaelf