Collaborative Micro-Filtering

I know I haven't been entirely forthcoming with what Vyoo is up to, but today's post will start to paint the picture. I came across a great blog posting today talking about a phrase for something I've been trying to explain about what we're doing. It was written by another San Franciscan, Sarah Cooper, and she calls the idea "Collaborative Micro-filtering". As you may, or may not know, collaborative filtering is the idea of using either a large set of data or many inputs of data to help filter information for users. What makes micro-filtering even more useful is that it uses your specific preferences of likes and dislikes to determine what types of data to present to the user.

Collaborative micro-filtering would take four steps (according to Sarah):
1. Identify people I know
2. Allow me to judge my similarity to each person, for a variety of topics
3. Make recommendations based on first and second order relationships between me and the people I know
4. Continually refine this similarity index for each person I know, based on our rating patterns over time

We're taking an approach that is somewhat similar to this by determining individual tastes and preferences for things. They are some vivid differences with what we're doing, but this is the best example, without writing a tell all of what we're up to.

Another interesting blog posting was written by Josh Porter, a social networking designer who discusses Ben Schneider's "Circle of Relationships".

We're attacking a very common problem on the internet today: How do you improve recommendation engines and really give people what they want.