Who are the most influential #NEfollowers?

Who are the most influential #NEfollowers?

Dollywagon has developed an Influence Engine that can pin-point the key leaders, influencers and idea creators in any online community or network.

The fantastic @canny_lass has suggested we use the Influence Engine to study the community on Twitter that’s using the #NEfollowers hashtag.

We thought this was a great idea and have resolved give it a try until we either get bored or you beg us to stop.

We’ve set up the Influence Engine to capture every interaction on Twitter between people using #NEfollowers. It works by deploying more than 40 network analysis algorithms to analyse a person’s success in getting their messages or Twitter @ID to be propagated in tweets that contain the #NEfollowers hash tag.

This will reveal the most important , significant or influential members of the #NEfollowers community on Twitter. And it knocks the socks off any flaky buzz-metrics analysis that you might be familiar with.

You can read more about the science behind the Influence Engine here.

Here’s our first image of the network – individual tweeters are represented by the circles, or ‘nodes’. The bigger the node the more network influence it has (only the most influential nodes have been labelled to keep things simple).

click to enlarge

The table below reveals our first Top 40 list of #NEfollowers. It’s based on just the latest 1500 tweets (due to Twitter API restrictions) that contain the hashtag. But as the days and weeks progress we’ll be able to collect every NE#followers tweet to gain a definitive picture of the community.

click to enlarge

At present the list of #NEfollowers is ranked by General Network Influence. For the graph-theory enthusiasts among you (and I know there are lots out there…) this is a simple blend of ‘Indegree Single’ and ‘Page Rank’ network metrics.

If your Twitter @ID appears in the list, here’s why it’s there. The ‘Indegree Single’ algorithm tells us how often you are personally referenced, named-checked or RT’ed by the rest of the #NEfollowers community. This is information is then ‘blended’ with data from our ‘Page Rank’ algorithm to tell us how much ‘network coolness’ you have.

For instance, if you’ve got loads of people picking up your tweets, that’s cool. But if all the people picking up your tweets (i.e. the people you hang out with) are ‘network dorks’, that makes you a lot less cool.

If you’re concerned by your apparent ‘network-dorkiness’, don’t worry. As we collect more tweets the picture will change and your network ranking may well improve. We’ll also try to introduce new measures of ‘network cool’ as we get more familiar with the community.

As we embark on this experiment we’d really appreciate your feedback. We’re a North East-based business with big ambitions and we’d love to hear your ideas about how we can make the Influence Engine better.

Please use the comments box below and we’ll do our best to respond to every message.

p.s. if you want to know how the Influence Engine weeds out spammers read this

7 People read this post and left a comment

  • Chris 10/06/2010

    Interesting, but I can’t help but think it’s also rather flawed.

    For example, let’s say some leading light in community says “Lets all post our website addresses!” and the rest of the community accordingly tweets their URLs, your algorithm isn’t going to pick that up if it’s only looking at personally references, named-checks and RTs. That person is clearly quite influential though.

    Similarly, if someone is adding the tag too *all* their tweets and the community choruses for them to stop with “@name please stop spamming”, your algorithm is going to push the spammer to the up the influence ranking when really their actions should be pushing them down.

    Without analysing the *context* of tweets, whether they represent a positive or negative weighting, I can’t see how this chart is particularly useful.

    It is pretty though, and that’s something.

  • Tweets that mention Who are the most influential #NEfollowers? | Dollywagon -- Topsy.com 10/06/2010

    [...] This post was mentioned on Twitter by Herb Kim, davidcoxon, Codeworks Connect, Kristian Lunde, Colin and others. Colin said: RT @Dollywaggon – Check out the complete Influence Engine Top 40 list of #NEfollowers here http://bit.ly/cbInFB – #mathswin [...]

  • deakaz 10/06/2010

    This is a great idea. I will be keeping a eye on this!

  • Colin (@colharris) 10/06/2010

    Fame at last! Thanks for providing this service

  • Sally Harris 10/06/2010

    Interesting criteria used for the analytics. Could all social networking media incl /FB/web links be aggregated into the results to create a joined up set of reports? Influential meets SE results?
    It’s facinating!

  • admin 10/06/2010

    We thought Chris has such a good question we’ve dedicated a whole new blog post to it that you can read here

  • admin 10/06/2010

    In response to Sally Harris’ question – we find it’s best to analyse each of the networks you mention independently. Then you can look for influential people who appear in one or more networks and potentially devise a way to ‘blend’ their results. Nice idea!

Leave a Reply