The Noisy Channel

January 13th, 2009 · 77 Comments · General

A few weeks ago, there was a flame war about Twitter authority, and I was all too eager to throw fuel on the pyre. But now that the blogosphere has calmed down a bit, I’d like to propose a ranking measure that I think might work. My apologies if it isn’t original. In fact, if you’ve seen it elsewhere, please point me to it.

• Influence(X) = Expected number of people who will read a tweet that X tweets, including all retweets of that tweet. For simplicity, we assume that, if a person reads the same message twice (because of retweets), both readings count.
• If X is a member of Followers(Y), then there is a 1/||Following(X)|| probability that X will read a tweet posted by Y, where Following(X) is the set of people that X follows.
• If X reads a tweet from Y, there’s a constant probability p that X will retweet it.

This model is obviously simplistic in all three assumptions. But I think it’s a reasonable first cut. In particular, it accounts for the inflation that occurs from people who follow in the hopes of reciprocity. There’s less value in being followed by someone who follows a lot of people, because that person is less likely to read your messages or retweet them.

Of course, there’s room for adding more realism to this model, but I hope it is at least close enough to the truth to be interesting.

From this model, it’s easy to measure someone’s influence recursively, assuming that we know the constant retweet probability p:

The recursion is infinite over a graph with directed cycles, but rapidly converges as high powers of p approach zero. I would think this measure wouldn’t be hard to compute to a reasonable accuracy.

This measure strikes me as a PageRank for Twitter or any system with similar properties. There’s more room for nuance, but I at least find this approach more plausible than the ones I’ve seen. It also strikes me as hard to game, since it isn’t counting retweets, and it’s hard to add much influence through followers who don’t have any influence themselves.

What do folks think? Has anyone tried this? If not, is there anyone who’d like to try hacking an application to compute it? Either way, please let me know!

77 responses so far ↓

• 1 WSDM 2010: Day 2 // Feb 6, 2010 at 12:01 am

[…] with modeling authority and influence in social networks, a problem in which I take a deep personal interest. Another inferred attributes of social network users based on those of other users in their […]

• 2 http://thenoisychannel.com/2009/01/13/a-… « AuthTweet // Feb 9, 2010 at 12:11 pm

• 3 FXPAL Blog » Blog Archive » Making sense of Twitter search // Feb 10, 2010 at 11:44 am

[…] to the results set; other sort orders such as the number of followers, recency of tweet, TunkRank, […]

• 4 Are Ashton Kutcher and Puff Daddy the Most Influential Twitter Users? // Mar 20, 2010 at 7:32 pm

[…] none of them model the scarcity of user attention, which is the motivating principle of TunkRank. Nor do they ground “influence” in any outcome external to […]

• 5 Feedburner and Friendfeed - Taking all credibility away from Feedburner's readership chicklet // Apr 22, 2010 at 9:34 am

[…] To me, it’s Google’s responsibility to intervene.  The company that expresses algorithmic prowess on so many complex patterns should have no trouble in doing so with blog engagement.  The raw numbers displayed in feedburner chicklets are no more reliable than the 1990s hit counters which allowed unscrupulous webmasters to “start off” with high numbers in order to mislead the readers that a site was popular.  Perhaps we need a pagerank for Twitter/Friendfeed followers. […]

• 6 FluidDB » Blog Archive » TunkRank scores added to FluidDB // May 5, 2010 at 6:34 pm

[…] and is maintained at tunkrank.com by Jason Adams. Given a Twitter user name, TunkRank computes a measure of that user’s influence based on followers of the user, how many people the user’s […]

• 7 TunkRank scores added to FluidDB // May 5, 2010 at 9:39 pm

[…] Pretty slick! To learn more about the TunkRank measure of Twitter influence / authority, check out this post. If you enjoyed this post, make sure you subscribe to my RSS […]

• 8 My Twitter Environment aka Twitter Tools I Use | Seasoned Ruminations // Jul 12, 2010 at 1:47 pm

[…] TunkRank http://tunkrank.com/ This is an adaptation of Google’s PageRank and measures inbound and outbound attention.  A much better description is available at: http://thenoisychannel.com/2009/01/13/a-twitter-analog-to-pagerank/ […]

• 9 FXPAL Blog » Blog Archive » Promoting TunkRank // Jul 16, 2010 at 10:51 am

[…] last year, Daniel Tunkelang proposed a way to measure people’s influence on Twitter; this metric was dubbed TunkRank, and Jason […]

• 10 Research award for microblog search « IRepeat — IR // Jul 23, 2010 at 1:55 am

[…] What criteria should we pursue to help people find useful information in microblog collections?  Surely time plays a role here.  Topical relevance is a likely suspect, as are various types of reputation factors such as TunkRank (and here). […]

• 11 The Online Influence Players | Ressive Networks // Oct 29, 2010 at 10:24 am

[…] For more on TunkRank check out this page and this blog. […]

• 12 Quora // Jan 23, 2011 at 5:43 pm

What are some existing “PeopleRank” algorithms?…

In Twitter context TunkRank by Daniel Tunkelang (http://thenoisychannel.com/2009/01/13/a-twitter-analog-to-pagerank/) is rather well-known. It has been implemented by Jason Adams (http://tunkrank.com/)….

• 13 Identifying Influencers on Twitter // Apr 16, 2011 at 7:52 pm

[…] course, I can’t help suggesting that TunkRank might be a more useful indicator than follower count. Unfortunately the authors don’t seem to […]

• 14 How to Measure your Twitter Influence | SEO in Simple Terms « Unclank // Jun 14, 2011 at 11:24 pm

[…] the algorithm behind TunkRank? Well it’s based on Daniel Tunkelang’s proposed Twitter influence algorithm (taken from The War on Attention Poverty: Measuring Twitter Authority) via seo-hacker.com […]

• 15 TunkRank, midiendo nuestra influencia en Twitter // Jun 27, 2011 at 2:19 pm

[…] que pretende medir la influencia de nuestra cuenta de Twitter. Así pues, TunkRank nace de un algoritmo similar al empleado por el pageRank de Google, de forma que nos devolverá un valor entre 0 y 100 para reflejar según este la […]

• 16 Don't Let Social Media Marketers Turn You into a Newt // Aug 3, 2011 at 10:20 am

[…] you might have, and there are even some people thinking seriously about how to measure that kind of influence.Another question that David raised was if there were wiser investments to be made.A figure that […]

• 17 Ranking Social Media Authority « geosocialite // Aug 26, 2011 at 8:18 am

[…] spamming techniques. Then, there was another approach by Daniel Tunkelang, first posted on his blog thenoisychannel, which got implemented as TunkRank. Unfortunately, at the time of this writing, I could not test […]

• 18 TunkRank: Twitter-Influence-Metric mit Chancen? // Oct 1, 2011 at 10:15 pm

[…] Startseite About TunkRank A Twitter Analog to PageRank TunkRank Improvements Wer ist Daniel Tunkelang How to Measure your Twitter Influence | SEO in […]

• 19 Social Media im Wandel: Ist das Glas halbleer oder halbvoll? // Oct 1, 2011 at 10:20 pm

[…] Startseite About TunkRank A Twitter Analog to PageRank TunkRank Improvements Wer ist Daniel Tunkelang How to Measure your Twitter Influence | SEO in […]

• 20 "How to weed out weak twitter users"?; who don't twit much or have very less followers. - Quora // Oct 5, 2011 at 6:23 am

[…] some of these pages: Simple explanation here: http://tunkrank.com/aboutMore detailed explanation: http://thenoisychannel.com/2009/…Slide Version: http://mendicantbug.com/2010/07/…– TunkRank: We measure influence on Twitter […]

• 21 PeopleRank: What are some existing “PeopleRank” algorithms? | 曹志士 // Feb 1, 2012 at 11:39 pm

[…] Kamil, Vilan Natanzon, Raza Rizvi, (more) In Twitter context TunkRank by Daniel Tunkelang (http://thenoisychannel.com/2009/&#8230😉 is rather well-known.It has been implemented by Jason Adams (http://tunkrank.com/). […]

• 22 Science as a Strategy // Apr 25, 2012 at 9:25 am

[…] set the stage, I told the story of TunkRank: how, back in 2009, I proposed a Twitter influence measure based on an explicit model of attention […]

• 23 Yannis // Jul 23, 2012 at 8:27 am

Hi Dan. I know this is quite a belated follow-up to your post, but I was wondering: as I(X) is the expected number of people who read something by X, how does the method cope with cases where after convergence, I(X) is greater than N, where N the #nodes in the network? In your implementation, do you impose a threshold N for each element of I per iteration?
Thanks and kudos for Tunkrank!

• 24 Daniel Tunkelang // Jul 23, 2012 at 8:49 am

Yannis, it doesn’t try to. While it’s theoretically possible for the model to exceed this threshold, it won’t happen with realistic parameter values. Indeed, the model doesn’t even try to prevent cycles, nor does it avoid double-counting if person reads the same message twice (because of retweets).

Lots of room to make the model more complex / realistic. But I figured it was best to keep it simple and understandable, at least to start off.

• 25 Information, Attention, and Trust | It's All About Business // Jan 2, 2013 at 5:29 am

[…] the allocation of attention by explicitly modeling attention scarcity. Influence measures like TunkRank recognize human attention as a finite quantity.Attention bond mechanisms fight spam by enabling […]

• 26 Tony // Jan 11, 2013 at 1:03 pm

I find this to be a very interesting experiment, though I do not tweet myself. Apologies for the delayed comment, my question is if “influence” is really captured by simply reading something (or calculating readers). For someone that read and retweeted, it seems like your tweet may have “influenced” them a bit more than a reader. Perhaps calculating readership is difficult enough that there is no benefit going beyond that. Would like to hear your thoughts.

• 27 Daniel Tunkelang // Jan 11, 2013 at 2:29 pm

Tony, that’s a fair point. Influence is a fuzzy concept, but I’m comfortable with any definition that involved getting people to spend a scarce economic good, like their attention.