Twitter recently open sourced an algorithm designed to ease the process of running recommendation engines at large scale. Called DIMSUM, the algorithm pre-processes pairs of possible matches so the other algorithms in the process don’t waste resources on poor choices.
Streaming music service Spotify has acquired The Echo Nest and its graph of musical data spanning more than 35 million songs and 2 million artists. It’s an easy way for Spotify to match companies like Google and Pandora on the data science front.
This is an interesting patent application, in part because of its techniques and in part because — like many technology-related patent applications — it’s hard to see how it’s particularly novel. The idea of using someone’s social graph to find influential connections that could inform mobile-app recommendations is pretty good, but at the core aren’t we just talking about the decision to value one variable more than another in a recommendation system?
Thanks to the rating systems in place on such popular websites as Yelp, Amazon and eBay, many people are comfortable evaluating things in absolute terms: a two-star restaurant, a B movie and so on. But new MIT research says this approach is fundamentally flawed.
The BBC wants to bring personal video recommendations to its iPlayer catch-up service, and it’s been testing a number of methods and algorithms as part of a multiyear research project. The broadcaster just finished a final field test, and its lessons learned are worth a look.
There was too much news today in the web video world, but here are the highlights! Babelgum has acquired Chad Vader, Next New Networks is launching a health and fitness network, Revision3 is now available via iPhone app and Michael Buckley no longer needs glasses.
Online product recommendation systems take a wide variety of approaches, but they can be shuffled into four main categories. Sites can boost sales and user loyalty by combining these categories to create the best overall experience.