1 Comment

Summary:

Goombah, the product of a Maine-based startup called Emergent Music, is announcing itself today. It’s another music discovery tool that uses collaborative filtering to recommend new tracks. Unlike the competition, which we’ve written about more than once, Goombah doesn’t run as an iTunes plug-in with a […]

Goombah, the product of a Maine-based startup called Emergent Music, is announcing itself today. It’s another music discovery tool that uses collaborative filtering to recommend new tracks.

Unlike the competition, which we’ve written about more than once, Goombah doesn’t run as an iTunes plug-in with a web-based manager, but instead is an over-sized 38.1 MB (on the Mac) download. It does show you other users’ entire music libraries, though, rather than just the overlaps.

By Liz Gannes

You're subscribed! If you like, you can update your settings

Related stories

  1. I am CEO of Emergent Music, the company that has developed Goombah. As Liz has pointed out, we have taken a very different approach to music recommendations. Goombah uses each user’s desktop CPU to process the large quantities of data necessary to produce very good recommendations. The 38.1 MB it takes on disk is mostly data that Goombah analyzes to calculate the best matching members for each user. We like to say that it creates the perfect word-of-mouth network for each user. It’s a lot of data because it takes into consideration the entire music collection of each user when searching for the best matches.

    Goombah also gives users the ability to do targeted searches by playlist, specific tracks and favorite artists — all modified by their chosen level of adventurousness.

    We believe that this approach results in recommendations of both people and music that are more accurate, interesting and full of surprises.

    Share

Comments have been disabled for this post