How the new Digg Recommendation Engine works

Davey Winder
Tuesday, 01 July 2008 06:07

Your IT - Home IT

Filtering the 16,000 new entries on Digg every day is an impossible task, for a human. However, the Digg Recommendation Engine can do it for you by analysing your past activity and revealing what Diggers Like You have been looking at...

Digg founder Kevin Rose has announced the availability of the long anticipated Recommendation Engine, putting paid to speculation that it was all a myth. The feature will start rolling out to registered users this week on a random basis. As soon as a red beta flag appears on the upcoming stories tab, the Recommendation Engine is active for your account.

Why do you need any such thing? Well that should be obvious to anyone who has used Digg in anger any time during the last couple of years. With 16,000 new stories being tagged every day on average, it is pretty much impossible to keep up with them all let alone keep on top of the quality ones.

Sure, there are numerous ways to filter the stories that appear on Digg, from the 'most popular' option to broad subject flagging. The trouble really starts, however, when you want to see what is new and interesting rather than just the latter.

For this most people will head for the Upcoming stories section, and that's where you will be confronted by recommendations by the bucket full. Which is where the Recommendation Engine comes in, to intelligently filter this vast pool of new stuff for you.

It does this by making use of the information that you have already given it. Namely, the stories that you have already Dugg yourself. This tells the Recommendation Engine a couple of things as Anton Kast, Digg's Lead Scientist, says "that you recommend the story to other users and, less obviously, that the users who Dugg the story before you are good at finding content."

Read how Diggers Like You bring the wisdom of small crowds into the process on the next page...

CONTINUED