Regulars will know that the Ferret is very partial to collaborative filter (also known as recommender) systems. These things use the power of other people’s recommendations, allied to your likes and dislikes, to help you find new stuff that you might enjoy.
I will go so far as to say that I believe that this technology will eventually become an indispensable part of the Internet, especially as the quantity of data grows and grows over the coming years. People already have enough trouble finding what they’re looking for now, just think how bad it will be when there’s ten times the amount of information on the Web.
Here’s a round up of some of the ones I’ve come across recently. If anyone knows of any more, drop me a line or comment here and I’ll add them to the list.
I’ve used MovieLens for a long time to help me locate movies I might like, and we’ve recently reported on the iLike service which is trying to do the same for the Web. Of course all of these services are only as good as the number of people contributing to them, but still… Anyway, here you go.
* MovieLens. Over a million ratings for just about every movie made in the recent past, and then some. The Unplugged version used to deliver recommendations to PDAs, but is currently defunct, more’s the pity. Part of the all-powerful GroupLens research project at the University of Minnesota.
* Libra. ‘LIBRA is a recommender system for books that learns a profile of your personal reading tastes from examples that you provide of books that you like or dislike. It then uses this profile to recommend other titles that it believes you will find interesting.’
* Jester. ‘Jester uses a collaborative filtering algorithm called EigenTaste to recommend jokes based on your ratings of a set of sample jokes.’
* iLike. Collaborative filter for locating sympatico web pages. Still in beta test. ‘Generally speaking I like! finds the users that have marked similar pages as you have and find [sic] other links that they liked the most.’
* iRate. Help with finding music tracks based on your personal tastes. As with many of these systems, currently very much a work in progress. ‘You rate the tracks it downloads and the server uses your ratings and other people’s to guess what you’ll like. The tracks are downloaded from websites which allow free and legal downloads of their music.’
* Scoop. A collaborative community web site builder. ‘Scoop is designed to enable your website to become a community. It empowers your visitors to be the producers of the site, contributing news and discussion, and making sure that the signal remains high.’
* StumbleUpon. More web surfing collaborative goodness. ‘We are a community-based, word-of-mouth approach to websurfing – pages you “stumble upon” come from like-minded people who share your interests.’
* Dooyoo & Epinions. Qualitative collaboration sites which offer a more commercialized version of the recommender service. Users contribute their reviews and opinions on a particular product, which helps others decide whether to buy one or not.
* Amazon. Stacked to the ceiling with recommendation services which are personalised to your buying habits. Slick commercial marketing end of the spectrum, and very professionally done.
* Gnod. The Global Network of Dreams. You tell it what you like and it will come back with new recommendations. Music, movies, books and web sites. ‘You might call it a search-engine to find things you don’t know about.’