How to use the Digg Recommendation Engine to your advantage

  • 0
  • July 1, 2008
Patrick Altoft

Patrick Altoft

Director of Strategy

Digg is about to make some fundamental changes to the way stories are displayed in the time between submission and either being made popular or expiring.

The changes mean that the stories displayed in the Upcoming section are stories recommended by your friends so any spam stories won’t get the eyeballs they do at the moment.

You can still view the old Upcoming section but as you can see in the screenshot below, this is pretty well hidden.

Digg Recommendation Engine

The key question for people wanting their stories to become more popular (everybody?) is – How can we make sure our stories are recommended to as many people as possible?

According to the white paper the algorithm looks at the number of stories you have in common with a particular Digger (i.e. the number of stories you both Dugg) and divides it by the total number of stories you or they Dugg.

This gives a correlation coefficient between 0 and 1 which shows how much you have in common with the other Digger.

Once the Engine has determined your Diggers Like You, your Recommendations consist of stories that your Diggers Like You have already Dugg, minus the stories you already Dugg or Buried.

There are some extra steps, like the diversity rules and the promotability constraint described below, but this is the basic idea.

If you monitor and Digg only the submissions of a particular top user then in theory you will end up being 100% correlated to that user and they should start to see stories you have Dugg in their Recommendations.

The best strategy is to figure out which users are actively Digging stories from the Recommendations section and make sure you Digg all the stories they Digg. If you see a user who only Diggs stories on the homepage then there is little point attempting to make sure your stories are featured in their Recommendations.

Free of charge. Unsubscribe anytime.