Heya,
I'm working on a collaborative filtering website and could use some test data to tweak my algo and see how it performs under load.
Concept: Basically, users come to the site and rate movies, authors, games or music. The system matches the user with other users with similar ratings and suggests items that these profiles rate highly. It is similar to what websites like netflix or amazon do, but substantially different in execution and results.
What I Need: Real-world profiles with ratings. I've tested as far as I can with friends and auto-generated data and I'd like to see how it performs in the real world so I can see if any performance or contextual problems emerge.
What Can You Do: Just register on the website: www.kindakarma.com. You only need to provide a username and password (no email or other personal information). And rate items that you love/hate. The system should suggest items to you, and get better at it as you rate more things. If you want to go above and beyond, you can post here or email me with your general sentiments about how well you like the suggestions the system gives you.
Any takers?
Much thanks,
Bob Jamison
bob@kindakarma.com
www.kindakarma.com
I'm working on a collaborative filtering website and could use some test data to tweak my algo and see how it performs under load.
Concept: Basically, users come to the site and rate movies, authors, games or music. The system matches the user with other users with similar ratings and suggests items that these profiles rate highly. It is similar to what websites like netflix or amazon do, but substantially different in execution and results.
What I Need: Real-world profiles with ratings. I've tested as far as I can with friends and auto-generated data and I'd like to see how it performs in the real world so I can see if any performance or contextual problems emerge.
What Can You Do: Just register on the website: www.kindakarma.com. You only need to provide a username and password (no email or other personal information). And rate items that you love/hate. The system should suggest items to you, and get better at it as you rate more things. If you want to go above and beyond, you can post here or email me with your general sentiments about how well you like the suggestions the system gives you.
Any takers?
Much thanks,
Bob Jamison
bob@kindakarma.com
www.kindakarma.com
Facebook
Twitter