The data and method. I used NodeXL to extract tweets. Going by the assumption that the most informative tweets will be the most popular (and therefore most retweeted) I selected tweets based on a set of simple "rules", weighted to achieve a balance of tweets over the 2 days. As commonly happens at conferences the volume of tweets and retweeting reduced over time. I selected those retweeted 8+ times to 1PM on first day, 5+ RTs thereafter on day 1, and 4+ RTs for day 2. I removed "spam" tweets (people outside the conference hall wanting to make a political point).

There were, overall, 1,922 tweets posted or retweeted during the week leading up to the conference, and the two days of the conference. These were posted by 425 tweeters, with 4,733 retweets.

NodeXL extracts data on posts tweeted during the period of study, plus posts retweeted during that period. As can be seen from the summary below, some of the pre-conference tweets will have received considerable attention prior to the conference.

The particular focus of this analysis is the tweets posted during the conference. This is a period where new relationships are made, new connections forged on social media, and also potentially useful summaries of key findings from the conference.

For the two days of the conference there were 1,804 tweets by 398 tweeters, with 3,149 retweets. The top 20 tweeters from these two days (by number of retweets) are listed below. For this period 87 tweeters (22% of tweeters) received 80% of retweets, while 125 tweeters (31% of tweeters) received no retweets.

First off, a bit of detail about data, methods and top tweeters. Scroll down for top tweets.

Items